Removal of sulfur and nitrogen compounds from gas oil by adsorption

A thesis submitted to the College of Graduate and Postdoctoral Studies

In partial fulfillment of the Requirements for the

Degree of Master in Science

In the Department of Chemical and Biological Engineering

University of Saskatchewan

Saskatoon, SK

Canada

By

Anakaren Botana de la Cruz

© Copyright Anakaren Botana de la Cruz, July 2020. All rights reserved

i

PERMISSION TO USE

In presenting this thesis in partial fulfillment of the requirements for a Postgraduate degree from

the University of Saskatchewan, I agree that the libraries of this University may make it freely

available for inspection. I further agree that permission for copying of this thesis in any manner,

in whole or in part, for scholarly purposes may be granted by Dr. Ajay Dalai and Dr. John Adjaye,

who supervised my thesis work, or in their absence, by the Head of the Department or the Dean of

the College of Engineering. It is understood that any copying or publication or use of this thesis or

parts thereof for financial gain is prohibited and shall not be allowed without my written

permission. It is also understood that due recognition shall be given to me and to the University of

Saskatchewan in any scholarly use which may be made of any material in my thesis. Requests for

permission to copy or to make other uses of materials in this thesis/dissertation in whole or part

should be addressed to:

Head of the Department

Chemical and Biological Engineering

University of Saskatchewan

57 Campus Drive

Saskatoon, Saskatchewan

S7N 5A9 Canada

OR

Dean

College of Graduate and Postdoctoral Studies

University of Saskatchewan

116 Thorvaldson Building, 110 Science Place

Saskatoon, Saskatchewan S7N 5C9 Canada

ii

ABSTRACT

Catalytic hydrotreatment is known worldwide as the process where concentrations of species like

nitrogen, sulfur, oxygen and various metals are reduced to acceptable levels in the refining process.

This process involves high temperatures and pressures, which results in major economic expense

to refineries. These process conditions are dependent on the nature of the petroleum feedstock to

be processed. Bitumen-derived gas oils from Athabasca oil sands can have high concentrations of

organo-sulfur (~ 4 wt.%) and nitrogen (~ 0.4 wt.%) species; thus, posing a major challenge to

downstream processing in refineries. Due to the challenge represented by these heteroatoms (S, N)

during conventional catalytic hydrotreatment, other methods that involve non-catalytic processes

using π-acceptor-immobilized adsorbents have garnered immense interest of researchers in this

field due to its low temperature and pressure requirements. Though, mixed metal oxides such as

mesoporous alumina with desirable textural properties (surface area, pore diameter and pore

volume) have the tendency to be used as adsorbent support for the immobilization of π-acceptor

moieties via a linker, their potential has been less explored. In this study, an attempt was made to

investigate the mesoporous alumina support-linker-π acceptor, charge-transfer complex (CTC)

structure for the adsorption of these heteroatoms present in the oil. This novel alumina-based

adsorbent, combined with the non-catalytic adsorption method, was effective to selectively remove

the refractory sulfur and nitrogen compounds from model feedstocks under milder conditions

(room temperature and atmospheric pressure) as compared to catalytic hydrotreatment.

The research work was divided in 3 phases. Phase 1 was focused on the synthesis and

characterization of the alumina-based adsorbents and their characterization. Three adsorbents were

synthesized using a charge- transfer complex (CTC) moiety consisting of a support, linker and π-

acceptor. Alumina was selected to be the supporting materials for its textural properties; three

different alumina- based supports were used: mesoporous alumina, titania-substituted alumina and

commercial alumina for comparison purposes. These three supports went under reaction to get a

linker, ethylenediamine (EDA) attached to them and then a πacceptor, 2,7-Dinitro-9-fluorenone

(DNF). The adsorbents and supports were characterized using BET, FTIR, TGA techniques to

ascertain their physicochemical properties. The extent of π-acceptor functionalization on supports

was characterized using XRD, TGA and XPS techniques. In phase 2, the adsorbents were

iii

examined for the desulfurization of a model oil feed with 500 ppm of sulfur. The commercial γ-

Al2O3-EDA-DNF adsorbent successfully removed 90.4 wt% of sulfur. The substitution of Ti in

the framework of mesoporous Al2O3 did not promote its desulfurization efficiency. The higher

desulfurization activity of the commercial γ-Al2O3 based adsorbent than that of others is attributed

to its textural properties.

Among the three adsorbents screened a commercial γ-Al2O3 CTC adsorbent (Adsorbent C) showed

the highest desulfurization in a short time period. Therefore, in Phase 3, the sulfur adsorption

isotherms and kinetic were examined. The kinetics of sulfur adsorption followed a pseudo-second-

order model with the CTC adsorbents. The regeneration of used adsorbent was studied with three

different polar solvents such as chloroform, dichloromethane and carbon tetrachloride.

Dichloromethane was found to be the most suitable solvent for extracting a major part of sulfur

compounds contained in the pores of the spent adsorbent. Thermodynamic parameters such as Ea,

ΔG, ΔH and ΔS provided a better insight into the adsorption.

iv

ACKNOWLEDGEMENT

I want to take the opportunity to thank everyone that supported me through my MSc. Program.

Firstly, I would like to express my gratitude to my supervisors, Dr. Dalai and Dr. John Adjaye for

giving me a once in a lifetime opportunity to pursue my master’s degree under their guidance and

for all the shared knowledge. I am extremely thankful for having the opportunity to achieve another

level of education under their supervision and constant support.

I thank my advisory Committee members, Drs. Amira Abdelrasoul and Lifeng Zhang for their

input during my research and their availability when it was needed. I will also like to thank Drs.

Misra Prachee, Philip Boahene and Sundaramurthy Vedachalam for their continuous support in

the laboratory and through my program. I will to extend my gratitude to Heli Eunike, Richard

Blondin, Dushmanthhi Jayasinge and Rosa Do for their assistance for analytical instruments’ use.

I would also like to thank fellow students in Catalysis and Chemical Reaction Engineering

Laboratories (CCREL) for making a support community among our research group. I thank my

friend and fellow student Dylan, with whom I have shared so many memories and personal

experiences.

Lastly, on a personal note, I would like to thank my parents, Araceli de la Cruz Cruz and Jorge

Alberto Botana Ponce and my brother Jorge Alberto Botana de la Cruz, for their constant support,

love and encouragement through the distance. You are my everyday motivation. I also want to

thank all my family back in Mexico who have always showed me their love and support. I thank

Jacob, for giving me the words of encouragement when I needed them and your constant love and

support to make hard days easier.

v

DEDICATION

This thesis is dedicated to my parents

Araceli de la Cruz Cruz and Jorge Alberto Botana Ponce

My accomplishments will always be dedicated to you

vi

Table of Contents

PERMISSION TO USE i

ABSTRACT ii

ACKNOWLEDGEMENT iv

DEDICATION v

TABLE OF CONTENTS vi

LIST OF TABLES viii

LIST OF FIGURES ix

NOMENCLATURE xi

ABBREVIATIONS xii

Chapter 1 : Introduction and Thesis Outline 1

1.1 Introduction 1

1.2 Knowledge gaps 6

1.3 Hypothesis 6

1.4 Research Objectives 6

1.5 Organization of the thesis 7

Chapter 2 : Literature review 8

2.1 Global energy consumption and oil reserves 8

2.2 Alternative non-catalytic methods 9

2.3 Materials used for nitrogen and sulfur impurities adsorption 10

2.4 Synthesis of mesoporous materials 16

2.4.1 Al2O3-based materials syntheses and adsorption performance with linker and π-

acceptor 22

2.5 Adsorption kinetics and thermodynamics 23

2.6 Regeneration and reusability 24

Chapter 3 : Synthesis, characterization and batch adsorption experiments 25

3.1 Abstract 25

3.2 Introduction 26

3.3 Experimental 28

3.3.1 Materials 28

3.3.2 Synthesis of adsorbents 28

vii

3.3.3 Characterization of supports and adsorbents 30

3.3.4 Procedure for adsorption experiments 31

3.4 Results and discussion 32

3.4.1 N2-adsorption measurement 32

3.4.2 Fourier transform infrared analysis 35

3.4.3 Thermogravimetric analysis 37

3.4.4 X-ray diffraction 38

3.4.5 X-ray photoelectron spectroscopy 39

3.4.6 CHNS elemental analysis 40

3.4.7 Adsorption experiments using ultra-low sulfur diesel (ULSD) feedstock 42

3.4.8 Effects of adsorption time 44

3.4.9 Effects of feed concentration on adsorption capacity 45

3.5 Conclusions 46

Chapter 4 : Optimization, regeneration and kinetic studies 48

4.1. Experimental methods 48

4.1.1 Materials/ Chemicals 48

4.1.2 Preparation of adsorbents 48

4.2 Process parameter Optimization 49

4.2.1 Adsorption parameters optimization studies 50

4.3 Adsorption kinetics 52

4.3.1 Kinetic studies 53

4.4.1. Adsorption thermodynamic studies 57

4.5.1 Regeneration and reusability studies 60

Chapter 5 : Conclusions and recommendations 62

5.1 Conclusions 62

5.2 Recommendations 63

References 64

Appendix A: Permission to reuse submitted paper Tables and Figures 73

Appendix B: Additional table of results from process parameter optimization studies and

activation energy (Ea) plots for adsorbents A, B and C 74

viii

LIST OF TABLES

Table 1.1 Adsorbents used and applications 3

Table 2.1 Feed characteristics of Bitumen derived gas oil, adapted from Yang et al., 2020. 8

Table 2.2 Different methods use for desulfurization and denitrogenation of petroleum

feedstock 10

Table 2.3 Various materials used as adsorbents for the removal of nitrogen and sulfur of

petroleum feed. 11

Table 2.4 Textural properties of the materials. Adapted from Baia et al., 2017. 12

Table 2.5 Adsorption capacities of alumina and clays towards nitrogen and sulfur compounds

based on Model 1: First Order 13

Table 2.6 Textural properties of the adsorbents. Adapted from Silveira et al.,2015. 17

Table 2.7 Characteristics of the mesoporous alumina samples. Adapted from Alphonse

et al., 2013. 18

Table 2.8 Textural properties from various Al2O3 supported materials. Adapted from

Badoga et al. 2014. 20

Table 3.1 Textural properties of the synthesized support, support-linker, and support-

linker-π-acceptor. 35

Table 3.2 CHNS elemental composition of supports, support + linker and final adsorbent 41

Table 4.1 Optimization parameters and their corresponding range used for the Central Composite

Design. 49

Table 4.2 Design of experiments for adsorption of sulfur and nitrogen species 50

Table 4.3 Linear equations for adsorbents A, B and C from Composite Design Expert

based on the proposed set of experiments. 52

Table 4.4 Kinetic parameters for pseudo-first and pseudo-second-order model 55

Table 4.5 Activation energy calculated values for adsorbents A, B and C. 57

Table 4.6 Free activation energy values for adsorbents A, B and C. 58

Table 4.7 Enthalpy and entropy values for adsorbents A, B and C. 58

ix

LIST OF FIGURES

Figure 3.1 Schematic of synthesized adsorbent for desulfurization 28

Figure 3.2 N2-adsorption/desorption isotherms for Al2O3,Ti-substituted meso-alumina and

commercial gamma-alumina their corresponding DNF-immobilized supports

(Al2O3-EDA-DNF and Ti-substituted meso-alumina -EDA-DNF). 33

Figure 3.3 Pore size distribution of the DNF immobilized support Al2O3-EDA-DNF, Ti-

substituted Al2O3-EDA-DNF and commercial alumina-EDA-DNF. 34

Figure 3.4 FT-IR spectra of the alumina-based supports 36

Figure 3.5 FT-IR spectra of adsorbents, ethylenediamine and dinitrofluorenone. 36

Figure 3.6 Thermal degradation profiles of adsorbents and dinitrofluorenone 37

Figure 3.7 X-ray diffraction for alumina supports. (A) Mesoporous alumina support, (B)

Ti-substituted meso-Al2O3 and (C) commercial γ-alumina supports. 38

Figure 3.8 X-ray diffraction for alumina based adsorbents and π-acceptor compound. 39

Figure 3.9 X-ray photoelectron spectroscopy of Adsorbent B. 40

Figure 3.10 Comparison of sulfur removal efficiency of Adsorbents A, B and C. Adsorption

parameters: Feed = 500 ppm of sulfur in ULSD, T = 22 °C, stirring = 400 rpm, atmospheric

pressure, adsorbent to feed ratio = 1:5, time = 24 h). 42

Figure 3.11 Sulfur and nitrogen removal efficiency of adsorbents A, B and C with ULSD non-

spiked feed. Adsorption parameters: T = 22 °C, stirring = 400 rpm, atmospheric pressure,

adsorbent to feed ratio = 1:5, time = 24 h. 43

Figure 3.12 Effects of adsorption time on sulfur removal (Adsorption parameters: T = 22 °C,

stirring = 400 rpm, atmospheric pressure, adsorbent to feed ratio = 1:5,

time = 0 to 96 h). 44

Figure 3.13 Effects of initial concentration of sulfur model feed on adsorption capacity of

adsorbents A, B and C (Adsorption parameters: T = 22 °C, stirring = 400 rpm,

atmospheric pressure, adsorbent to feed ratio = 1:5, time = 24 h). 46

Figure 4.1 The three-dimensional response surfaces: effects of temperature and adsorbent

loading on desulfurization activity of adsorbents A, B and C. 51

x

Figure 4.2 Pseudo first-order adsorption kinetics of adsorbents A, B and C. Temperature =22°C,

feed to adsorbent ratio=5. The solid line is the results for the kinetic model; the dash line is

pseudo first-order-model simulation 54

Figure 4.3. Pseudo first-order adsorption kinetics of adsorbents A, B and C. Temperature =22°C,

feed to adsorbent ratio=5. The solid line is the results for the kinetic model; the dash line is

pseudo first-order-model simulation. 55

Figure 4.4 Schematic of Soxhlet extraction set-up used for regeneration. 59

Figure 4.5 Sulfur removal from Adsorbent A after 8 h of regeneration with chloroform, carbon

tetrachloride and dichloromethane 60

Figure 4.6 Reusability results for adsorbent A using dichloromethane as solvent 61

Figure A.1 Permission to use published article, tables and figures for Chapter 3 73

Figure B.1 Optimization result table for adsorbent A 74

Figure B.2 Optimization result table for adsorbent B 75

Figure B.3 Optimization result table for adsorbent C 76

Figure B.4 Activation energy (Ea) plot ln KD vs 1/T (K-1) for adsorbent A 76

Figure B.5 Activation energy (Ea) plot ln KD vs 1/T (K-1) for adsorbent B 77

Figure B.6 Activation energy (Ea) plot ln KD vs 1/T (K-1) for adsorbent C 77

Figure B.7 Free activation energy, enthalpy and entropy plots for adsorbents A, B and C 78

xi

NOMENCLATURE

F-127 tri-block copolymer

FT-IR Fourier-transform infrared spectroscopy

HPLC high performance liquid chromatography

keV kiloelectronvolt

Kv kilovolt

nm nanometers

P-123 tri-block copolymer

SBA-15 Mesoporous alumina

SDAs structure directing agents

ZSM-5 aluminosilicate zeolite

ΔG0 free activation energy

ΔS0 entropy

ΔH0 enthalpy

xii

ABBREVIATIONS

BET Brunauer-Emmett-Teller

BJH Barrett-Joyner-Halenda

BN basic nitrogen

CTC charge transfer complex

DCM dichloromethane

DNF 2,7 – dinitro – fluorenone

EDA ethylenediamine

HDN hydrodenitrogenation

HDS hydrodesulfurization

HGO heavy gas oil

LGO light gas oil

MDF model diesel fuel

NBN non-basic nitrogen

PGMA Poly (glycidyl methacrylate)

PPD Process parameter design

ppm parts per million

ppmw parts per million

PTFE Polytetrafluoroethylene

rpm revolutions per minute

TGA Thermogravimetric analysis

ULSD ultra low sulfur diesel

XPS X-ray photoelectron spectroscopy

XRD X-ray diffraction

1

Chapter 1 : Introduction and Thesis Outline

1.1 Introduction

As the population of the world increases and the impact of industry, transportation, greenhouse

emissions and other pollutants grow, emissions from combustion of fuels are being regulated

(Yoosuk et al., 2020). Due to this situation, more effort is being put towards delivering and

meeting the energy demand while meeting environmental regulations. Current oil reserves in the

world place Canada as number 7 with most of their oil reserves contained in the Athabasca, Alberta

region (Speight, 2020). Naturally, this oil deposits are in the form of oil sands, a combination of

sand, clay, water and bitumen. That bitumen is then refine and transform in a variety of fuels;

many different processes are involved in order to achieve the quality of fuels required (Mccabe,

2020). Among the processes that are involved in the refining process, hydrotreatment is an

important one since its where impurities like sulfur, nitrogen, oxygen and aromatics are removed.

The removal of these impurities happens in presence of excess hydrogen, high temperature and

pressure (260-427˚C and up to 13 MPa) using a catalyst typically CoMo/Al2O3 or NiMo/Al2O3.

Sulfur removal (Hydrodesulfurization (HDS)), Nitrogen removal (Hydrodenitrogenation (HDN)),

saturation of aromatics (Hydrodearomatization (HAD HDA)) and metals removal

(Hydrodemetallation (HDM)) are some of the reactions that take place during a typical

hydrotreating process (Satterfield 1996).

HDS is the catalytic process where sulfur is removed from petroleum products; organic sulfur

species are converted to sulfur dioxide (H2S) after reacting with the excess hydrogen (H2) present

in the system (Bu et al., 2011). The reaction for HDS is shown below, where R represents the

hydrocarbon units contained on the hydrotreatment feed.

(1.1)

The concentration and nature of the sulfur organic compounds changes over the boiling range.

Thiols, sulfides and disulfides, thiophenes, benzothiophenes, dibenzothiophenes are the major

sulfur compounds present in gas oil (Song, 2003). On the other hand, nitrogen in the gas oil

"[R-S]" + 𝐻2 → "[R-H]" + H2S

2

feedstock is present predominantly as heterocyclic aromatic compounds. Non-basic (five

membered ring) and basic (six-membered ring) are the two types of aromatic nitrogen compounds

present predominantly in the Athabasca bitumen (Fu et al., 2006). The reaction for HDN is shown

below:

(1.2)

Where R represents the hydrocarbon chains contained on the hydrotreatment feed. Some of the

nitrogen compounds present in the bitumen-derived gas oil are indole, carbazoles, indoline,

acriline and aniline (Fu et al., 2006). As it was mentioned before, the government has set the

maximum content of sulfur in fuels of 10 parts per million (ppm); concentration that will be

achieves after hydrotreating the feed (Yitzhaki et al., 1995). In order to explain the importance of

nitrogen removal an insight to the catalyst used on the hydrotreatment process is needed. A catalyst

is a substance that in a small amount causes a large change as it increases the rate of reaction

toward equilibrium without being appreciably consumed in the process. The rate of reaction

depends on pressure, temperature, concentration of reactants and products, and other variables

(Satterfield, 1996).

However, the catalyst may undergo through some major changes in its structure and composition

as a result of its participation in the reaction. Catalysts are usually synthesized using a support with

textural properties such as high surface area, pore size and pore volume (Bu et al., 2011). They can

also used different metals to act as promoter and enhance the activity during the reaction; however,

it may lose its activity by a variety of different reasons. Poisoning is the presence of elements such

as phosphorous (P), arsenic (As), silica (Si), sodium (Na), vanadium (V), tungsten (W) and/or iron

(Fe). Another phenomenon that can occur is fouling, reduction of the active area by sintering or

migration and the loss of active species (Satterfield, 1996).

There are some challenges on processing high N and S compound concentration on gas oil, where

crude oils with more than 0.5%wt sulfur need to be treated extensively during petroleum refining.

The distillation process segregates sulfur species in higher concentrations into the higher-boiling

fractions and distillation residues. Removing sulfur from petroleum products is one of the most

"[R-N]" + 𝐻2 → "[R-H]" + 𝑁𝐻3

3

important processes in a refinery to produce fuels compliant with environmental regulations. Basic

nitrogen compounds are particularly undesirable in crude oil fractions as they deactivate the acidic

sites on catalyst. Some nitrogen compounds are also corrosive. An alternative to prevent catalyst

deactivation are non-catalytic methods to remove the sulfur and nitrogen present in gas oil (Song

et al., 2003). Some of the materials that have been used as adsorbent and its application are shown

in Table 1.1.

Table 1.1 Adsorbents used and applications

Adsorbent Applications

Silica Gel Drying of gases, refrigerants, organic solvents, transformer oils

Desiccant in packings and double glazing ( Ziolo et al., 1995; Ullah

et al., 2013)

Activated

Alumina

Removal of HCl from hydrogen (Fleming et al.,1988; Ullah et al.,

2013)

Removal of fluorine in alkylation process (Heilig 1994; Weber et al.,

1991)

Carbons Removal of SOX and NOX (Mochida et al., 2000; Sumathi et al.,

2009, 2010)

Purification of helium (Chang and Wu, 2009; Favvas et al., 2015)

Clean-up of nuclear off-gases (Munakata et al., 1999)

Zeolites Oxygen from air (Li et al., 1998)

Drying of gasses (Auerbach et al., 2003; Li et al., 2008)

Functionalized

Polymers and

resins

Water purification (Akelah and Sherrington, 1981; Puoci, et al.,

2008)

Recovery and purification of steroids, amino acids (Hentze and

Antonietti, 2002)

Separation of fatty acids from water and toluene (Kale et al., 2015)

Clay Treatment of edible oils (Bokade and Yadav, 2009; Richardson,

1978)

Removal of organic pigments (Espantaleón et al., 2003; Zhou et al.,

2014)

Refining of mineral oils (Vural, 2020; Wu et al., 2017)

The focus of this study will be adsorption, defined as the adhesion of molecules (adsorbates) in an

extreme thin layer to the surfaces of solid bodies (adsorbents) with which they are in contact (Baia

et al., 2017). It has also been widely used due to less operational cost, specific target and high

efficiency removing undesired compounds. The adsorption process is generally classified

as physisorption (characteristic of weak van der Waals forces and hydrogen bonding)

or chemisorption (characteristic of covalent bonding). Physical adsorption resembles

4

the condensation of gases to liquids and depends on the physical, or van der Waals, force of

attraction between the solid adsorbent and the adsorbate molecules. Chemical adsorption is

adsorption in which the forces involved are valence forces of the same kind as those operating in

the formation of chemical compounds (Nathan and Scobell, 2012).

The technique here proposed is charge-transfer complex (CTC) which occurs between a π-acceptor

compound functionalized with a support and π-donor compounds that can be found on the gas oil

(Macaud et al., 2004). The functionalized particles consist of support, linker and π-acceptor. The

support would bring the mesoporous surface to the linker to be attached to, using a linker could

have an effect on the selective adsorption of NBN compounds and lastly, the π-acceptor would be

the electron poor donor compounds contributing to the selective adsorption of π-donor compounds

present in the oil (Rizwan et al., 2013).

The support in an inner substance that helps spreading out an expensive catalyst onto its surface

such as Pt, it can also help stabilize the activity towards the adsorption process. It may be used in

pellet or powder form (Maity et al., 2003). Also, in a similar process are the catalyst, which are

promoted with active metals. The presence of an active metal onto the support, might improve the

textural properties of the support, leading to an increase of activity towards the adsorption process

(Wachs, 2005).

Activated alumina is a mesoporous material that has high surface area: 100 to 600 m2/g, its pore

size distribution its mesoporous in nature, it is a low cost material when compared to titania (Ti)

and it has shown thermal stability (Badoga et al., 2017). On the other hand, Ti helps in NOx

reduction emissions in hydrotreating, is more acidic than alumina, which enhances the hydrogen

sulfide (H2S) formations (affinity towards sulfur compounds). Metal oxide supports contain alkali,

alkaline and noble metals, also have been widely employed as solid catalysts; either as active

phases or supports (Wachs, 2005). Both are use for their acid-base and redox properties and

constitute the largest family of catalysts in heterogeneous catalysis.

The function of the linker is to immobilize the π-acceptor onto the mesoporous supports (Misra,

2017). Amines, hydrazines and hydroxylamine have been used as linkers and have been reported

in the literature. The linker allows the immobilization to take place on the support without

substituting an electron with-drawing group (on the support) with an electron-donor compounds,

compounds present in gas oil (Sévignon et al. 2005).

5

Previous reported studies have calculated theoretically (Milenkovic et al. 2004) and demonstrated

(Meille et al. 1998; Milenkovic et al. 1999; Sévignon et al. 2005) that symmetrical π-acceptor

compounds based on polynitrosubstituted 9-fluorenones have showed selectivity towards sulfur

compounds such as dibenzothiophenes, benzothiophenes and aromatic compounds. As mentioned

before, the sulfur compounds present in the oil contain electron rich aromatic systems; this

characteristic might enhance the affinity of the adsorbate towards the adsorbent and create an

attraction to facilitate the adsorption of the sulfur compounds. Based on the available literature,

there are many adsorbents that have been tested based on the CTC complex formation, but to the

best of our knowledge, the meso- porous alumina-based supports with the CTC complex formation

has not been reported for desulfurization. Mesoporous alumina seems to be good support as it has

a high surface area, pore diameter and pore volume for anchoring the π-acceptor. In this study,

2,7-dinitro-9- fluorenone (DNF) functionalized pristine mesoporous Al2O3, Ti-substituted

mesoporous Al2O3 and commercial γ-Al2O3 were prepared, characterized and evaluated in the aim

to treat a model feed for desulfurization. Afterwards, kinetic studies and thermodynamic

parameters were calculated to bring an insight into the adsorption taking place. Lastly, regeneration

and reusability of adsorbent A was tested using solvent removal method.

6

1.2 Knowledge gaps

1. The performance of π-acceptor functionalized particles for the adsorption of nitrogen and sulfur

compounds is limited in the literature. The synthesis of functionalized mesoporous Al2O3 and Ti-

incorporated Al2O3 supported adsorbents via π-acceptor and their application for sulfur and

nitrogen removal is limited in the literature.

2. Regenerability, optimization and kinetic studies of mesoporous Al2O3 based adsorbents are

scarce in the literature.

1.3 Hypothesis

1. Post-treatment of processed oil using non-catalytic methods may successfully remove nitrogen

and sulfur compounds, using mesoporous Al2O3 support and π-acceptor may increase selectivity.

Incorporating TiO2 onto mesoporous Al2O3 support would increase the adsorption capacity.

2. Optimization, kinetics and regeneration studies will determine the suitability of these adsorbents

for liquid feedstocks containing sulfur and nitrogen impurities

1.4 Research Objectives

To study the performance of π-acceptor immobilized alumina-based materials as adsorbent for the

removal of sulfur and nitrogen impurities from liquid fuels. The goal is to improve the efficiency

of hydrotreating in terms of less catalyst loading, lower reaction temperature, and severity of

hydrotreating conditions (T, P, space velocity, hydrogen/oil ratio…)

Sub-objectives:

To investigate the effects of the proposed synthesis method and its impact on the adsorbent

properties for sulfur and nitrogen compounds present in liquid fuels.

To study process optimizations, kinetics and long-term stability of potential adsorbent

identified in above sub-objective.

7

1.5 Organization of the thesis

This thesis contains 5 chapters covering detailed synthesis, characterization, experimental testing

and conclusions from all the phases of the research. Chapter 2 focuses on the literature review

starting with the Global energy consumption and oil reserves, followed by the alternative non-

catalytic methods, materials used for nitrogen and sulfur impurities adsorption. Synthesis of

mesoporous materials and Al2O3-based materials and their adsorption performance with linker and

π-acceptor. Lastly, adsorption kinetics, thermodynamics and regeneration were further studied.

Chapter 3 presents the synthesis of alumina-based adsorbents followed by the characterization and

batch adsorption experiments performed using a diesel model feed. Chapter 4 describes the process

optimization studies, regeneration experiments, kinetic study of the adsorption process and the

thermodynamics of the process.

Chapter 5 summarizes the key findings from this work and provides recommendations for possible

future work in this area.

8

Chapter 2 : Literature review

This chapter discusses the review of the literature related to this research. Non-catalytic methods

have been discussed as an alternative due to its simplicity as compared to hydrotreatment. Among

many different methods, adsorption seems to be the most widely studied. However, its success

depends on finding an ideal adsorbent that can effectively remove nitrogen and sulfur species from

liquid fuels.

2.1 Global energy consumption and oil reserves

The world depends on different energy resources to function. The demand for oil started to increase

after World War II and correlated to the rapid growth of the world´s economy.

The main energy sources used worldwide are oil (32 %), coal (26%), gas (23%), biomass (10%),

electricity from hydro and nuclear (9%) and heat (<1%). The United States dominated the world

production until other countries started their own production. Currently, 60 % of the world´s oil

production are from mostly six countries (Yang et al., 2020).

Canada´s oil reserve is in 7th place with 4 % of the global oil reserve (Ha et al., 2018). These

reserves are mostly contained in the oil sands in Cold Lake and Peace River regions of Northern

Alberta. Alberta´s oil sands contain around 166 billion barrels of oil (Yang et al., 2020), which by

nature contain more minerals, metals and other polluting compounds. These oil reserves are known

as oil sands due to their high concentration in sand, clay and water as compared to those present

in crude oil. It contains up to 80-85 % of sand and clay, 4-6 % of water, and 10-12% of bitumen.

Table 2.1 Feed characteristics of Bitumen derived gas oil, adapted from Yang et al., 2020.

Parameter Light Gas

Oil (LGO)

Heavy Gas

Oil (HGO)

Sulfur content

(ppm)

29000 38000

Nitrogen content

(ppm)

1700 4000

Three major stages are used for converting crude oil to fuels. The stages can be referred to as

separation, conversion and treating, and involve processes such as distillation, cracking, reforming

9

and hydrotreating just to mention a few. The Canadian government has set the sulfur content in

gasoline to be 10 ppm (Ha et al., 2018), therefore, special consideration has to be taken when

designing processes for feed with high sulfur concentration to meet the environmental regulations.

Hydrotreating is the process in which contaminants in the hydrocarbons are removed in the

presence of excess hydrogen (H2) using high temperature (350-400˚C) and pressure (5-10 MPa).

Table 2.1 shows the amount of sulfur and nitrogen compounds typically contained in LGO and

HGO feeds; sulfur content varies from 0.3-5 wt.% and nitrogen from 0.1-0.8 wt In the presence of

a catalyst, sulfur and nitrogen compounds react with H2 to be removed as hydrogen sulfide (H2S)

and ammonia (NH3), respectively. The vehicular emission from fuel combustion can contribute

greatly when the sulfur species in fuels is converted to sulfur oxides after combustion; this can

ultimately lead to acid rain. Nitrogen compounds present in oil, on the other hand, also contribute

to the nitrogen oxide emissions (NOx) and causes other problems in refineries such as catalyst

deactivation, coke formation, and they can be found to be poison for acid functionality in catalysts

(Satterfield, 1996).

2.2 Alternative non-catalytic methods

Extensive research (Bu et al., 2011; Li et al., 2015; Liu et al., 2020; Misra et al., 2018; Qi et al.,

2020; Zolotareva et al., 2019) has been done over the last years in an effort to find different non-

catalytic methods other than hydrotreating. As it was mentioned before, high H2 consumption,

high temperature and pressure used in hydrotreating process can be costly; thus impacting on the

cost of fuels (Li et al., 2009). Among different technologies that can be used for reducing sulfur

and nitrogen concentrations in refineries feedstock, the non-catalytic methods are gaining

significant attention. There are many advantages when using non-catalytic methods, low

temperatures and pressures are uses as compared to the conventional hydrotreating methods.

(Zolotareva et al., 2019) has recently reported Ionic liquids (IL) as an effective method due to its

low temperature and pressure; they can also be more selective than some solvents such as acetone,

pyrimidinone, polyalkylene glycol, imidazolidinone, dimethyl sulfoxide and polyethylene glycol

that are found to have high volatility, flammability and toxicity (Toor et al., 2013). Bedda et al.,

2019, reported the DS of light cycle oil using extraction solvents such as acetonitrile, N-

methylpyrrolidone, and N,N-dimethylformamide. The performance of the solvent was evaluated

and it was concluded that a solvent/feed ratio from 1.0 to 2.5 has a higher sulfur removal as the

10

ratio increased. Acetronitile removed 53.3 ppmw%, N-methylpyrrolidone 74.8 ppmw% and N,N-

dimethylformamide removed 73.3% from the total sulfur content of initially 2157 ppmw. Table

2.2 shows different methods used with advantages and disadvantages for comparison purposes.

Table 2.2 Different methods use for desulfurization and denitrogenation of petroleum feedstock

Method Advantages Disadvantages

Adsorption

(Mckav and Ireland, 1985;

Misra et al., 2015a;

Sampanthar et al., 2006)

Require less material.

Process can be carried at

room temperatura.

Control of the properties

and surfaces.

Solid applications.

Non-catalytic method

Large amount of

adsorbents needed for

industrial aplications.

Regenerability needs to be

studied.

Absorption

(Li et al., 2009; Wachs,

2005)

Liquid and gas applications.

Not necessarily require

elevated pressure.

Conversion of a hazardous

chemical to a safe

compound.

Extensive areas of liquid

surfaces in contact with

gas phases.

Not enough information

for further comparison.

Ultrasound assisted

oxidative desulfurization

(Liu et al., 2020; Zolotareva

et al., 2019)

Efficient desulfurization

method

Complements the HDS

process.

Only focus on sulfur

compounds.

Widely study in the

industry

Based on the literature review, adsorption was found to be a more suitable method for the removal

of sulfur and nitrogen compounds from liquid fuels. Several adsorbents have been reported in the

literature, most of them are described in the following.

2.3 Materials used for nitrogen and sulfur impurities adsorption

The aim of this section is to provide an extensive literature review on what has been done on the

adsorption of nitrogen and sulfur compounds from liquid fuels. Many different materials have been

used as adsorbents, such as zeolites, activated carbon, silica, alumina, ion exchange resins, titania

nanotubes just to mentioned a few. Based on the characteristics of the materials, alumina was

11

found as the more suitable adsorbent support, having a relatively high surface area from 100 to

600 m2/g, being mesoporous in nature, thermally stable and having a relatively low cost compared

to titania, which has shown to have high efficiency for nitrogen and sulfur removal (Badoga et al.,

2017; Maity et al., 2003).

Some of the different materials used as adsorbents are mentioned in Table 2.3. Many different

adsorbents and feed materials have been tested, however, there is limited information on the

adsorption of sulfur and nitrogen compounds from model and real oil fractions using alumina-

based adsorbents.

Table 2.3 Various materials used as adsorbents for the removal of nitrogen and sulfur of

petroleum feed.

Baia et al. 2017, studied the adsorption of sulfur and nitrogen compounds contained in a straight

run diesel feedstock using three different types of clay, designated as A, B and C. Their study

Method Feedstock

Adsorption

capacity (mg

S,N /g

adsorbent)

Temperature

(°C) Highlights Reference

Activated

carbon

Real diesel-

398 ppmw (S)

Model diesel

– 400 ppmw

(S)

77mg/g, 88

mg/g 25, 30, 75

Polycyclic

aromatics and

sulfur removal

(Bu et al.,

2011)

Ion

exchange

resins

Model feed

(benzene and

thiophene)

Not stated.

Reduction of

97% of sulfur

120

Effective for

basic nitrogen

species

removal

(Yang et al.,

2004)

Copper

zeolites

Gasoline- 335

ppmw (S)

Diesel- 430

ppmw (S)

8.89 and 13.12

mg S/g

adsorbent

23

Effective

adsorption for

sulfur

compounds

(Hernández-

Maldonado

and Yang,

2003)

Titania

based

nanotubes

Model diesel

blends

14.6 mg S/ g

adsorbent 20

Adsorption of

nitrogen and

sulfur

compounds

(Rendon-

Rivera et al.,

2016)

12

included kinetic and isotherm experiments. A commercial alumina was used for comparison

purposes due to its properties of adsorbing many different substances. The textural properties of

the alumina and clays are shown in Table 2.4. Straight run diesel was used as a feed with a

composition of 3906 mg kg-1 of sulfur and 522 mg kg-1 of nitrogen. Kinetic tests were performed

in a Dubnoff reciprocal shaking bath, using a shaking frequency of 150 rpm at 40°C. Seven flasks

containing 10 mL of diesel and 2 g of adsorbent were used and samples at 10, 30, 60, 120, 180,

240 and 300 minutes were analyzed.

Table 2.4 Textural properties of the materials. Adapted from Baia et al., 2017.

During the adsorption experiments, it was observed that the rate of adsorption was high during the

first 10 minutes, thereafter, it decreased. This was attributed to the large number of pores that are

available at the beginning of the process as compared to those available after a determined period

of time. By measuring the solution concentration at the equilibrium point, the authors noted that

the adsorption process in the liquid phase involved a competition between the solvent and the

solute.

As we can see from Table 2.5, Clay B showed better performance than the other materials in both,

nitrogen and sulfur adsorption capacities. This can be attributed to the presence of Brөnstead acids

and higher specific surface area than the other materials. In addition, Clay B and Clay C showed

different adsorption steps for both sulfur and nitrogen compounds, this particular characteristic can

be noticed on their respective adsorption-desorption isotherms. In another study carried by Sano

et. al. (Sano et al., 2004), where the surface area of various activated carbons was the selection

criteria for its expected performance, straight run gas oil (SRGO) was fed into a tube containing

activated carbons and activated fibers. Among the materials that were used for these experiments,

Pore volume (cm3 g-1)

Sample Specific area

(m2 g-1)

Mesoporous Microporous Pore

diameter

(nm)

Alumina 357 0.34 0.02 5

Clay A 95 0.45 0.01 22

Clay B 198 0.21 0.01 6

Clay C 273 0.30 0.02 9

13

MAXSORB-II showed the largest adsorption capacity for nitrogen species in gas oil. Even though

MAXSORB-II and MAXSORB-III both showed equally large surface areas (> 2000 m2/g) which

is considered a key characteristic for adsorption capacity, MAXSORB-II had a better performance

(Sano et al., 2004).

Table 2.5 Adsorption capacities of alumina and clays towards nitrogen and sulfur compounds

based on Model 1: First Order

Adsorbent qe, cal (mol·kg-1) nitrogen qe, cal (mol·kg-1) sulfur

Alumina 0.058 0.059

Clay A 0.074 0.018

Clay B 0.125 0.174

Clay C 0.116 0.126

Other materials, such as zeolites have been used as adsorbents as well with a π-complexation

structure (Yang et al., 2001). Yang and co-workers reported the adsorption process took place in

a fixed-bed reactor operated at ambient temperature and pressure. This chemical complexation has

been barely utilized. Even though there are some studies in the literature that mention this model,

few of them have achieved remarkable removal of nitrogen and sulfur compounds. Added to the

π-complexation characteristic of this material, during the adsorption process, a thin layer of AC

was used and showed that it can significantly increase the sulfur adsorption capacities of the π-

complexation sorbent. This copper-based zeolite is highly hydrophilic but no other properties of

this particular zeolite are mentioned. However, the fixed-bed adsorption results showed that the

sulfur contents of those commercial fuels used (gasoline and diesel) were drastically decreased.

In a different study by Santos et al. (Santos et al., 2012), diesel was used as feed for an adsorption

pre-treatment with the aim to remove nitrogen and sulfur compounds. Commercial silica-alumina

samples were calcined and then impregnated with various mixed metal oxides such as cerium,

nickel and molybdenum. The commercial name of the silica-alumina material is SIRAL 40 and is

composed by 60:40 Al2O3/SiO2 mass ratio. First, this material was calcined under atmospheric

pressure for 4 h at 823 K. Then, aqueous solutions containing the metal oxides were prepared for

wet impregnation of the parent material with Ni (NO3)3·6H2O, (NH4)6Mo7O24·4H2O or Ce

(NO3)3·6H2O. The concentration was aimed to be 0.17 mol of metal per 100 g of adsorbent, this

14

wet impregnation was carried out in a rotary evaporator at 353 K for 5 h. Finally, the material was

dried and calcined once more. For the adsorption experiments, 10 mL of the commercial diesel

containing 229 mg of total sulfur and 196 mg of total nitrogen per kg of diesel were mixed with

the adsorbent in different weights. Using 0.50 g, 0.75 g, 1.00 g, 1.50 g, 2.00 g and 2.50 g at 313 K

under stirring, the temperature was raised after 45 minutes to 343 K and kept constant for 45

minutes and then it was cool down. After this pre-treatment, the adsorption experiments were

carried out for 7 h.

The increase on the adsorption capacity per area of the materials can be attributed to the

incorporation of metal oxides. Molybdenum species showed the best performance, however a

reduction in the specific surface area was observed. An increase on the adsorption capacity was

not observed on a mass basis; therefore, it can be assume that incorporating molybdenum is

promising as long as there is a control on the loss of the surface area of the material.

Another type of material that has been studied for pre-treating oil feed are titania nanotubes.

Rendon-Rivera et al., 2016, used TiO2 nanoparticles and TiO2 nanotubes as adsorbents to study

their selective adsorption properties on liquid phase with sulfur and nitrogen organic compounds.

Using dibenzothiophenes (DBT), 4, 6-dimethyl DBT (4, 6-DMDBT), pyrrole and quinoline which

are compounds that represent some of those ones contained in diesel fuels. Model diesel blends

and these compounds were mixed to emulate a real diesel blend. Adsorption isotherms for the

compounds were determined at ambient temperature and pressure and fitted into Langmuir or

Tempkin models. To establish the influence of TiO2 nanoparticles transformation into nanotubes,

the commercial version of this material was used as the reference adsorbent. For the batch

adsorption experiments, four different model diesel fuels (MDF) were prepared. The adsorption

was carried out in a batch system at room temperature (20°C) at atmospheric pressure. TiO2 was

added in different quantities but keeping the model diesel fuels volume constant, then, the mixtures

were shaken using a magnetic stirrer to reach the equilibrium phase (typically, 0.5 h). After that,

some liquid was extracted, it was passed through PTFE syringe filters and it was stored in vials.

The fixed bed adsorption experiments were carried out at ambient conditions as well, using titanate

nanotubes or anatase as particles and only two of the MDF. A gas column was packed with 2 g of

adsorbent of between 80-100 mesh and placed in a special furnace. The materials were heated to

300°C prior to adsorption experiments for 3 h in the presence of nitrogen as a pretreatment. After

15

this, the adsorbent was allowed to cool down to room temperature and then hexane was used to

remove any entrapped gas.

In addition, another study from the same group were another type of zeolites, Cu-Y and Ag-Y, as

selective sorbents for the desulfurization of liquid fuels (Li et al., 2009). Compared to Na-Y, which

is one of the sorbents presented in the previous study Takahashi et al., 2002, Cu+-Y was prepared

by ion exchange zeolites with Na-Y, followed by reduction, the same process described in their

previous paper (Munakata et al., 1999). Molecular orbital computational technique was used to

determine the details on the materials. The bonding energies of the materials were calculated based

on the following equation:

𝐸𝑎𝑑𝑠 = 𝐸𝑎𝑑𝑠𝑜𝑟𝑏𝑎𝑡𝑒 + 𝐸𝑎𝑑𝑠𝑜𝑟𝑏𝑒𝑛𝑡 − 𝐸𝑎𝑑𝑠𝑜𝑟𝑏𝑒𝑛𝑡−𝑎𝑑𝑠𝑜𝑟𝑏𝑎𝑡𝑒 (2.1)

Where Eadsorbate is the total energy of thiophene; Eadsorbent is the total energy of the bare adsorbent

and Eadsorbent-adsorbate is the total energy of the adsorbate/adsorbent system. E is measured in

kcal/mol. Isotherms of benzenes and thiophene on Ag-Y and Cu-Y were fitted by the Langmuir-

Freundlich isotherms and then compared. These sorbents showed more adsorption capacity

towards thiophene and benzene than Na-Y at lower pressures than 10-3 atm and nearly the same

amounts at high partial pressures. Even though, the paper states that this is due to the π-

complexation with Ag and Cu, further research was done to confirm these theoretical findings. In

order to understand the strengths of this complexation between Ag and Cu. The neutron activation

analyses of the sorbent samples showed that the Ag exchange was 100% but the Cu exchange was

23%. On a per-cation basis, the π-complexation with Cu was stronger than the one with Ag. Results

showed that 0.92 thiophene molecule per Cu was obtained at 2X10-5 atm at 120°C, whereas only

0.42 thiophene molecule/Ag was obtained.

Hernández-Maldonado et al., 2003, studied the desulfurization of commercial liquid fuels by π-

complexation with Cu(I)-Y Zeolite. The experimental section was developed in a fixed bed

adsorber operated at ambient temperature and pressure, and the results yield to a concentration of

below the limit of detection of the instrument used, Flame photometric detection (FPD). As

mentioned in the previous papers by Takahashi et al., 2002, and Yang et al., 2001, molecular orbital

16

calculations have shown that the π-complexation bonds between Ag or Cu and thiophene are

stronger than those ones with benzene, this sorbents are suitable for a selective desulfurization of

transport fuels. The procedure was the same as previous sorbents, initially the adsorbents were

loaded inside the adsorber and heated in presence of helium, and then a sulfur freed hydrocarbon

feed was passes through the sorbent at a rate of 0.5 cm3/min. From this paper there is only so little

that can be discussed. Most of the key findings have already been mentioned in the two previous

papers; however, they also conducted a study with a guard bed reactor. When used, Cu(I)-Y

showed higher adsorption capacities at both, breakthrough and saturation points.

Velu et al. 2002, conducted a study using zeolites on the selective adsorption for removing sulfur

(SARS). The aim of this study was to explore some selective-zeolite based materials for removing

organic sulfur compounds from transportation fuels. In this research, commercial zeolites were

ion-exchanged with transition metal ions using three to five time the excess amount of a

concentration of 0.1 M metal nitrates at 80°C for 24 h. After this, the suspension was filtered,

washed with deionized water and dried at 80°C overnight and then calcined at 450°C for 6 h in air

atmosphere. After the preparation of these materials, adsorption took place in a batch reactor using

JP-8 jet fuel as the feed. In a 100 mL round bottom flask, 1 g of ion-exchange zeolite and 6 g of

feed were stirred at 80°C for 4-5 h. The treated jet fuel was then separated from the adsorbent and

analyzed by a GC equipped with a pulsed flame photometric detector (PFPD) using

dibenzothiophenes as a standard. The study showed that those zeolites ion-exchanged with Cerium

and Palladium, CeY-Zeolite and PdY-Zeolite respectively, removed the highest amount of sulfur,

60 and 58 wt.%, respectively. The species that were found in the untreated jet fuel were 2, 3-

DMBT and 2, 3, 7-TMBT. The sulfur concentration of the feed was around 736 ppm and higher

selectivity for the adsorption of 2, 3-DMBT.

2.4 Synthesis of mesoporous materials

In a different study by Silveira et al. 2015, the influence of nickel, cerium, molybdenum and cobalt

oxides impregnated on silica-alumina was evaluated for removing nitrogen and sulphur

compounds from a hydrotreated fuel stream. Silica-alumina with 40 wt. % of SiO2 was used as a

support for the materials used in this study. This support was calcined at 550°C for 4 h before it

was used to confirm the presence of metal oxides. The material was sieved to a particle size range

of +100 to 325 mesh. The addition of nickel, cerium, molybdenum and cobalt was carried out by

17

wet impregnation method resulting on a 2.5 wt. % present in the silica-alumina support. Adsorption

experiments were carried out in a 1:20 adsorbent to fuel ratio kept at 40°C for 45 minutes under

stirring at 150 rpm. Then the temperature was raised up to 70°C and kept for 45 minutes, and then

the temperature was reduced to room temperature and maintained for 15 hours. Specific surface

area and pore volume are fundamental characteristics that help in the explanation of the adsorption

performance. This study concluded that the addition of metal oxides can decrease the surface area

up to 25%.

As shown in Table 2.6, an approximate of 25% of the support area was lost after the wet

impregnation of metal oxides; however, there was not a significant reduction in the mesoporous

volume. The adsorption performance of these materials was measured by mmol of sulfur or

nitrogen removed per adsorbent area, where molybdenum was better than the other metal oxides.

In this particular case, a higher percentage of nitrogen was removed than that of sulfur. Several

alumina synthesis methods are available in the literature (Alphonse and Faure, 2013a, 2013b;

Badoga et al., 2015; Khodabandeh and Davis, 2014; Niesz et al., 2005; Yuan et al., 2008). The

general synthesis is based on a sol-gel process associated with a non-ionic block copolymer as

templates using ethanol as a solvent (Yuan et al., 2008). These mesoporous aluminas have large

surface areas (around 400 m2/g), pore volumes (around 0.70 cm3/g) and narrow pore size

distributions. Compared to silica, alumina is more used in the petroleum sector as catalysts

supports.

Table 2.6 Textural properties of the adsorbents. Adapted from Silveira et al.,2015.

Non-siliceous materials used surfactants as structure-directing agents (SDAs) or by using the nano-

casting method. According in previous reports (Alphonse and Faure, 2013a; Niesz et al., 2005),

the hydrolysis behavior of alumina is very complicated and it is significantly affected by acid,

water, temperature, humidity and other factors giving arise to different conditions for ordered

Adsorbent Specific surface area

(m2/g)

Mesoporous volume

(cm3/g)

Silica-alumina 451 0.84

Silica-alumina Mo 329 0.79

Silica-alumina Ce 328 0.80

Silica-alumina Ni 352 0.80

Silica-alumina Co 341 0.80

18

mesoporous aluminas. Among the different Al precursors, the most commonly used is the

aluminum isopropoxide. Different acids such as HNO3 and HCl are also added as the pH adjustors

for the hydrolysis of the precursors. The use of a polymer as a surfactant combined with an acid

has shown that large surface area alumina supports can be developed. Surfactants such as P-123

and F-127 were tested using different acid molar ratios. Narrow pore size distributions are obtained

whereas after calcination they tend to decrease to some extent. These supports have shown high

thermal stability up to 1000 °C. These three characteristics combined, result in enhancing the

potential of these mesoporous alumina and their applications. Also, Niesz et al., 2005, reported a

different synthesis method, that applies the same concept as Yuan et al. 2008. This material has a

similar arrangement of that of SBA-15, known as one of the groups of ordered mesoporous silica

materials. During this synthesis, two solutions were prepared, one containing the surfactant (P-

123) in a solvent (ethanol), and the second one containing the alumina precursor (aluminum tri-

tert-butoxide) and the acid (HCl). This mixture was left for aging at 75°C for 4.5 days, followed

by vacuum drying at 70°C for 12 h prior to calcination at 600°C for 5 h. Hydrolysis steps is the

key to achieving mesoporous morphology of the material. Also, using small amounts of water and

acid showed that this hydrolysis process slows down, so the right amount of water and acid is

crucial for this process to happen in a control rate and hence the formation of the mesoporous

structure of the material. The BET surface areas resulting from this process are shown in Table

2.7.

Table 2.7 Characteristics of the mesoporous alumina samples. Adapted from Alphonse et al.,

2013.

Sample names BET surface area

(m2/g) Pore volume (cm3/g) Pore diameter (Å)

MP-Al2O3_2 a 206 0.44 39-150

MP-Al2O3_6 a 410 0.80 67

MP-Al2O3_12 a 349 0.62 39-90 a the numbers in the samples stand for the [H2O]:[Al2O3] ratios.

Considering that surface characteristics of the supporting material are key for the adsorption

performance, many different researchers have tried to develop a synthesis method that results in

having large surface area, large pore volume and tunable pore size distribution. For instance,

19

Alphonse et al., 2013, synthesized alumina-based materials with the above mentioned

characteristics by adding triblock copolymers. In the procedure, 185 mL of hot water (85°C) was

added fast to 25.3 g of aluminum-tri-sec-butoxide, under vigorous stirring. After the aluminum

was dissolved into the water, 0.474 mL of nitric acid (HNO3, 68%) was added and kept for stirring

at 85°C for 24 h. As a last synthesis step, an additive was incorporated to the mixture and kept for

stirring at room temperature during 24 h and the material was calcined at 500°C for 24 h. Since

this particular study incorporated different additives and nitrates, there was a wide range of surface

areas and pore volumes varying from 354-501 m2/g and 1.53-2.63 cm3/g. Also, Gonz and Sastre

2001 obtained samples of mesoporous aluminas with a similar synthesis procedure. In their study,

non-ionic surfactants were used in different ratios and the alumina source was aluminum sec-

butoxide. First, the surfactant was dissolved in sec-butanol and aluminum sec-butoxide. Then, a

solution containing water, dipropylamine and half of the required sec-butanol were added. The

resulting mixture was stirred for 3 h, and then cooled to room temperature. After this, the sol-gel

solution was filtered, washed with ethanol and dried at 40°C for two days. The dried gel was heated

at 95°C for 6 h. The surfactant was removed by Soxhlet extraction with ethanol for 15 h prior to

drying of the solid at 40°C for 2 days followed by calcination. The obtained materials had large

surface area varying from 229 to 534 m2/g, due to different calcination temperatures. It was

observed that the lower the calcination temperature (550°C), the higher the surface area. This was

not the case for the pore volume, which showed little relationship with calcination temperatures.

The methodology for this research will aim to obtain a novel material to use for the adsorption of

nitrogen and sulfur species from bitumen derived gas oil.

As alumina has been a widely studied (Biswas et al., 2011; Tursiloadi et al., 2004) support material

for catalysts, so has other metal oxides incorporated in alumina (Badoga et al., 2017; Maity et al.,

2006; Vosoughi et al., 2017). Studies (Badoga et al., 2014; Guevara et al., 2008; Herbert et al.,

2005; Silva et al., 2015) have shown that addition of other metals such as TiO2 have increased the

activity around 3-5 times in hydrodesulfurization (HDS) process. So, in the aim to experiment with

different approaches than the ones available in the literature, a mixed metal oxide will be used as

support for the adsorbents in this study. There are certain characteristics that the support and the

metal oxide need to have, such as, active metal dispersion, metal-support interaction and the

textural properties shown by the support material (Badoga et al., 2014). Based on a study by

Badoga et al., 2014, 10 wt. % TiO2, Zr and Sn were supported on Al2O3 for a catalytic reaction.

20

Even though this study is not adsorption based, the synthesis of the materials was the focus of

attention due to the material properties. In Table 2.8, the textural properties of some of the

materials synthesized on this research are shown.

Table 2.8 Textural properties from various Al2O3 supported materials. Adapted from Badoga et

al. 2014.

Material BET surface area,

m2/g Pore volume, cm3/g Pore diameter, nm

ϒ-Al2O3 275 0.80 7.6

Mesoporous Al2O3 320 0.57 5.8

TiO2- Al2O3 480 0.86 5.2

Taking into consideration that surface characteristics of the supporting material are key for the

adsorption performance, many different researchers have tried to develop a synthesizing method

that results in having large surface area, large pore volume and tunable pore size distribution. One

of many synthesis methods available in the literature is from Alphonse and Faure, 2013. In their

work alumina based materials with the above mentioned characteristics where synthesized by

adding triblock copolymers. In the procedure, 185 mL of hot water (85°C) were added fast to 25.3

g of aluminum-tri-sec-butoxide, under vigorous stirring. After the aluminum source was dissolved

into the water, 0.474 mL of nitric acid (HNO3, 68%) were added and kept for stirring at 85°C for

24 h. As a last synthesis step, an additive was incorporated to the mixture and the resulting mixture

was stirred at room temperature for 24 h; the material was then calcined at 500°C for 24 h. Since

this particular study incorporated different additives and nitrates, there was a wide range of surface

areas and pore volumes, ranging from 354-501 m2/g and 1.53-2.63 cm3/g, respectively.

Also, Gonz and Sastre 2001, obtained samples of mesoporous aluminas with a similar synthesis

procedure. In this study, non-ionic surfactants were used in different ratios and the alumina source

was aluminum sec-butoxide. First, the surfactant was dissolved in sec-butanol and aluminum sec-

butoxide. Then, a solution containing water, dipropylamine and half of the required sec-butanol

were added. The resulting mixture was stirred for 3 h and then cooled down for 24 h at room

temperature. After this, the sol-gel solution was filtered, wash with ethanol and dried at 40°C for

two days. After this, the dried gel was heated at 95°C for 6 h in an open propylene bottle; the

21

surfactant was removed by Soxhlet extraction with ethanol for 15 h. then, this solid was dried for

48 h at 40°C prior to calcination for 4 h at 550°C. Part of this material was calcined a second time

in an air oven at 600 and 700°C for 3 h. The resulting materials had large surface areas ranging

from 229 to 534 m2/g; this is considering the different temperatures the samples were calcined. It

was observed that the lower the calcination temperature (550°C), the higher the surface area. This

was not the case for the pore volume, which showed little relation to the different calcination

temperatures. Another alumina synthesis procedure performed by Khodabandeh and Davis, 2014,

even though the application was for an alumina catalyst, the material preparation is not that

different from the previous ones already discussed. Aluminum sec-butoxide (1 g) was hydrolyzed

in water (15.10 g), followed by stirring for 1 h. A solution of sodium dodecylbenzenesulfonate

(0.49 g) in formamide (5.05 g) was added to the mixture. It was aged for 5 min and then heated at

110°C for 48 h. The solid was recovered by filtration, washed with deionized water and then dried

at room temperature. Several alumina precursors are used and based on their molecular

composition the correspondent materials for the procedure are calculated. Only the support

preparation and characterization will be discussed. The support properties are not thoroughly

discussed in this paper, however, the preparation method is not that much different from the ones

available in the literature (Alphonse and Faure, 2013a; Khodabandeh and Davis, 2014). According

to the authors, performing the synthesis with different ethanol/formamide/water solvents and

ethanol/water showed that it plays an important role for the material to have good surface

properties. However, other factors like synthesis time and temperature may also affect the final

properties of this one.

Niesz et al., 2005, reported a synthesis method for ordered mesoporous alumina using different

templates as structure directing agents. Even though several papers have been published regarding

the use of templates, until 2005, none of them was about the synthesis of ordered mesoporous

alumina. The paper here discussed presents a reproducible method for synthesizing mesoporous

alumina with an SBA-15-like arrangement of ordered channels, high surface area and narrow pore

size distribution. As a surfactant, Pluronic P123 was used, 1 g was dissolved in 12 ml of ethanol

and stirred for 15 minutes at 40°C, and this was named solution A. At the same time, solution B

was prepared with different amounts of hydrochloric acid (37 wt. %) and 6 mL of ethanol. Then,

2.46 g of aluminum tri-sec-butoxide was added to solution B slowly under vigorous stirring, the

22

two solutions were mixed together and continued to be stirred at 40°C. This homogeneous solution

was then left for aging in a Teflon container for three days at 40°C under N2 flow.

2.4.1 Al2O3-based materials syntheses and adsorption performance with linker and π-

acceptor

The technique here suggested, the support + linker + π-acceptor, is the formation of a charge

transfer complex (CTC), which occurs between a π-acceptor compound (immobilized on a

support) and π-donor compounds (that can be found in the gas oil). These could be basic nitrogen

(BN) and non-basic nitrogen (NBN) compounds. The support would provide the porous surface

for the attachment of the linker. The contribution of the linker (linear diamines) to the adsorbent

system will be dependent on the length of the amine, which could potentially influence the

selective adsorption of the NBN compounds. In this regard, the CTC formation could be affected

by how far the π-acceptor is extended from the support. The length on this linker may be the factor

that allows more interaction between the adsorbent and adsorbate (Chitanda et al., 2015). In this

study three different linkers where tested. Using a polymeric support, polyglycidyl methacrylate

(PGMA), followed by linker and then the π-acceptor compounds, four different materials were

synthesized. The results showed that the length of the linker had little to no effect on the loading

during the adsorption experiments. The adsorbent that showed a better performance compared to

the others was that one with diaminopropane as linker, followed by diaminobutane. There was no

significant difference between the other two adsorbents reported in the paper (Abedi et al., 2015).

The π-acceptor characteristic is present among several nitrogen compounds, these compounds are

electron poor due to the withdrawing of an electron caused by the nitro groups; therefore, these

compounds can act as π-acceptors. This is relevant due to the affinity or selectivity they can have

towards non basic nitrogen compounds, that are known to have a lone pair of electrons in their

nitrogen atoms localized in the aromatic system (Abedi et al., 2015). In contrast, the basic nitrogen

compounds have the lone pair of electron located on the nitrogen atom. Hence, these π-acceptor

compounds are major contributors to enhancing the adsorbents selectivity towards nitrogen

compound (both basic and non-basic). Compounds with characteristics to act as π-acceptors

include 2, 4, 7-trinitro-9-fluorenone (TriNF) and 2, 4, 5, 7-tetranitro-9-fluorenone (TENF); hence

there in different adsorbent materials for desulfurization and denitrogenation reactions. Attention

has been brought to metal oxides, particularly alumina, which is going to be used in this research

23

as the support material for the adsorbents to be developed. Many different synthesis methods of

alumina supports have been developed. Abedi et al., 2016, modified alumina support with P-

toluene sulfonic acid powder, linker and π-acceptor compounds dissolved in dioxane. The resulting

alumina-polyglycidyl methacrylate-co-ethylene glycol dimethacrylate polymer (Al-PGMA-DAP-

TENF) and its alumina-free derivative (PGMA-DAP-TENF) were synthesized, characterized and

evaluated for the sulfur and nitrogen removal performances. FT-IR spectra show that the

characteristics peaks of C-O and C=O groups are both represented, as well as the attachment of

the π- acceptor by the appearance of stretching frequencies at 1361cm-1. Alumina particles were

stable between 30°C to 800°C as confirmed by TGA technique. Al-PGMA-DAP-TENF particles

were decomposed at lower temperature and the amount of residue after 800 °C was lower than that

in case of Al-PGMA, which indicates a successful grafting and the formation of a new compound.

The performance of Al-PGMA- DAP-TENF was found to be superior to that of PGMA-DAP-

TENF as it removed about twice as much nitrogen compounds as other polymers did in LGO, and

HGO feeds. Both polymers removed more sulfur compounds in HGO feed than in LGO feed. This

is due to the high sulfur content of HGO as compared to that of LGO feed (Misra et al., 2015a).

Based on these properties, mesoporous Al2O3 and TiO2- Al2O3 synthesis were explored. As a

reference is commercial γ- Al2O3 which properties can compete to the ones of the materials

synthesized.

2.5 Adsorption kinetics and thermodynamics

To test and prove the capacity of an adsorbent, adsorption capacity is calculated. In the adsorption

process, the liquid (adsorbate) is exposed to the solid phase (adsorbent) and the distribution of

adsorbate-adsorbent depends on the affinity of the adsorbent towards the adsorbate. Adsorption

capacity is the amount of adsorbate that the adsorbent takes per unit mass of the adsorbent; usually

mg/g (El Qada et al., 2006). This adsorption equilibrium capacity is achieved when the rate the

molecules attached onto the adsorbent are equal to the rate at which they desorb from it (McKay

et al., 1985). Along with the adsorption capacity, adsorption isotherms are often determined in

order to study the adsorption performance of an adsorbent (Lin et al., 2007). Establishing

equilibrium is key to compare the adsorbent behavior for different adsorption systems

24

quantitatively; this parameter helps to design the adsorption system and calculate other adsorption

parameters.

Kinetic models provide insight to adsorption using equations to fit experimental behavior (Lin and

Wang, 2009). Fitting the kinetic data and discuss the possible mechanisms of adsorption based on

the model parameters are the best way to provide insight into the adsorption process (Pan and

Xing, 2010).

Furthermore, thermodynamic parameters may also provide an insight into the type of adsorption

taking place. Activation energy (Ea) can be calculated to determine if the adsorption process is a

diffused-controlled process (physical adsorption) or a chemical adsorption (Sismanoglu and Pura,

2001). To calculate Ea, the rate constant parameter k2 for pseudo-secondo order model widely used

to describe adsorptive behavior in different materials can be applied to the Arrhenius equation

(Chiou and Li, 2002; Foo and Hameed, 2010; Sismanoglu and Pura, 2001; Tang et al., 2012). In

addition, more thermodynamic parameters can be calculated based on the adsorption experiments

such as Gibbs energy, enthalpy and entropy, also reported to be calculated for adsorption systems

(Foo and Hameed, 2010; Schirmer, 1999; Tang et al., 2012; Zhang et al., 2017).

2.6 Regeneration and reusability

Among materials used as adsorbents, regeneration studies have been studied using mostly two

common methods: thermal treatment and solvent extraction (Almarri et al., 2009; Fei et al., 2017;

Shan et al., 2008; Yoosuk et al., 2020). Reusability of the material is important as it can provide

information for the industrial application of the process. Considering that in this work, we are

studying a charge-transfer complex (CTC) adsorbent, the π-acceptor characteristic becomes a

factor when selecting the solvents to use for regeneration. Ideally, an electron donor type of solvent

should be used since its polarity can successfully clean the pores of the adsorbent (Misra et al.,

2018); the remaining solvent in the adsorbent can be removed afterwards.

25

Chapter 3 : Synthesis, characterization and batch adsorption experiments

A similar version of this chapter has been published as a research article:

Botana de la Cruz, Anakaren, Philip Boahene, Sundaramurthy Vedachalam, Ajay K. Dalai, and

John Adjaye. 2020. “Adsorptive Desulfurization through Charge-Transfer Complex Using

Mesoporous Adsorbents.” Fuel 269, 117379.

Contribution of the MSc candidate:

Adsorbent synthesis, characterization, testing of the materials and data analysis was done by

Anakaren Botana de la Cruz. Anakaren Botana de la Cruz and Dr. Misra Prachee designed

synthesis method; adsorption experiments and set-up were designed by Anakaren Botana de la

Cruz. Anakaren Botana de la Cruz with suggestions and reviews from Drs. Ajay Dalai, Philip

Boahene and Sundaramurthy Vendachalam did all the manuscript writing and revision work.

Contribution of this chapter in the overall MSc. Work:

This part of the research work was focused on the synthesis of the functionalized adsorbents,

characterization and adsorption of sulfur model feed of the materials here developed.

3.1 Abstract

Mesoporous supports with a π-acceptor were developed for adsorptive desulfurization of

petroleum distillates. Pristine mesoporous Al2O3 and Ti-substituted mesoporous Al2O3 were

synthesized and used as supports along with commercial γ- Al2O3 for immobilizing

ethylenediamine (EDA) linker followed by the π-acceptor, 2, 7- Dinitro-9-fluorenone (DNF). The

adsorbents and supports were characterized using BET, FTIR, TGA techniques to ascertain their

physicochemical properties. The extent of π-acceptor functionalization on supports was

characterized using XRD, TGA and XPS techniques. Adsorbents were examined for the

desulfurization of a model oil feed with 500 ppm of sulfur. The commercial γ- Al2O3-EDA-DNF

adsorbent successfully removed 90.4 wt% of sulfur. The substitution of Ti in the framework of

mesoporous Al2O3 did not promote its desulfurization efficiency. The higher desulfurization

26

activity of the commercial γ- Al2O3 based adsorbent than that of others is attributed to its textural

properties.

3.2 Introduction

Refineries perform hydrodesulfurization (HDS) and hydrodenitrogenation (HDN) of petroleum

feedstock using NiMo/γ- Al2O3 and CoMo/γ- Al2O3 catalysts at the temperatures and pressures

ranges of 360-420°C and 8 to 9 MPa with excess supply of hydrogen for the removal of sulfur and

nitrogen species. The conventional HDS process is ineffective for the removal of refractory sulfur

compounds such as dibenzothiophenes and alkyl substituted dibenzothiophenes. Typically, 96-

98% of HDS activity is obtained by using the conventional hydrotreating technology; however,

the remaining 2-4% sulfur removal is difficult to achieve due to the presence of refractory sulfur

species (Lemaire et al., 2002). In addition, the refractory sulfur compounds are responsible for

catalyst deactivation and poisoning (Jayaraman et al., 2006). Tremendous efforts have been made

to improve the HDS process for desulfurization of refractory compounds by altering the process

parameters including temperature, pressure, and liquid hourly space velocity. Feed quality, which

is one of the three major factors in hydrotreating process efficiency, depends on the origin and

nature of crude oil. Heavy gas oil derived from Canadian oil sands contains about 4 wt.% sulfur

as compared to ~ 2.5 wt.% sulfur present in the oil shell formations. Canadian government limited

the sulfur content of gasoline and diesel to 14 and 15 ppm, respectively for the on-road vehicles

(www.canada.ca). The refractory sulfur compounds present in heavy gas oil should be desulfurized

to get ultra-low sulfur fuels that are required to comply with the regulations. The production of

ultra-low sulfur fuels by HDS requires a higher reactor temperature, pressure, as well as an

increased hydrogen rate. As the severe HDS increases fuel production costs, refineries are

developing and testing new HDS catalyst formulations by changing the active metals, promoters

and catalyst supports to reduce the cost of deep-desulfurization.

Highly efficient desulfurization technologies are required for the selective removal of refractory

sulfur compounds. Extensive research was performed in the area of alternate technologies

including oxidative desulfurization and adsorption (Almarri et al., 2009; Hernández-Maldonado et

al., 2003; Kim et al., 2006; Shan et al., 2008). Among the different techniques, adsorption has

shown many advantages such as relatively lower temperatures and pressures when compared to

the conventional HDS process. However, for this method to be effective, the development of the

27

adsorbents is key. In the literature, several materials were used as adsorbents for the removal of

sulfur and nitrogen species from gas oil. Velu et al., 2002, used Y-zeolite-based adsorbents for

selective desulfurization of jet-fuel. Activated carbon applications as adsorbent have also been

reported by Sano et al. 2004. Using a distilled fraction feed from an atmospheric distillatory, the

carbon-based supports were packed into a stainless-steel tube and were put in contact by pumping

the distilled fraction feed. This study suggested that adsorbent surface properties, as large surface

area and pore size might be important factors for better adsorption efficiency. Another highly used

material for desulfurization is silica-alumina. Silveira et al., 2015, reported silica-alumina

impregnated with different metal oxides. The loading of mixed metal oxides on silica-alumina

promoted the sulfur adsorption. In another study, Sarda et al. 2012, used Ni and Cu loaded alumina

and ZSM-5 supports for removing sulfur from a commercial diesel containing 325 ppm total sulfur.

This study concluded that the sulfur removal strongly depended on the nature of the metal, its

amount and the supporting material.

Recently, charge-transfer complex (CTC) adsorption has been reported for desulfurization (Abedi

et al., 2016; Hernández-Maldonado et al., 2003; Macaud et al., 2004; Misra et al., 2016; Song,

2003; Yang et al., 2001). This involves the formation of electron-donor acceptor complexes

through the interaction of aromatic π-donor compounds and π-acceptor molecules on support. The

refractory sulfur compounds of petroleum fuels contain electron-rich aromatic π-systems. They

can be adsorbed with electron-deficient systems through the CTC adsorption. The technique here

studied involves adsorptive desulfurization through a charge-transfer complex. Mesoporous

supports were functionalized with 2,7- Dinitro-9-fluorenone (π-acceptor) to create a sorbent that

binds sulfur compounds (π-donor compounds) via charge-transfer complexation. The π-acceptor

is electron-deficient and is known to attract heterocyclic sulfur compounds due to their electron-

rich nature. In this study, the π-acceptor is bound to mesoporous supports with the help of

ethylenediamine, a linker. The CTC complex formation is shown in Figure 3.1. Based on the

available literature, there are many adsorbents that have been tested based on the CTC complex

formation, but to the best of our knowledge, the mesoporous alumina-based supports with the CTC

complex formation has not been reported for desulfurization. Mesoporous alumina seems to be

good support as it has a high surface area, pore diameter and pore volume for anchoring the π-

acceptor. In this study, 2,7-dinitro-9-fluorenone (DNF) functionalized pristine mesoporous Al2O3,

28

Ti-substituted mesoporous Al2O3 and commercial γ- Al2O3 were prepared, characterized and

evaluated for desulfurization of a model feed.

3.3 Experimental

3.3.1 Materials

Poly (ethylene oxide)- block poly (propylene oxide) - block poly (ethylene oxide) (P123) was

purchased from Sigma Aldrich, Canada. Nitric acid (≥70.0%), methanol (>99.0%), toluene

(>99.0%) and acetic acid (>99.0%) were purchased from Fisher Chemical, Canada. Aluminum

isopropoxide (98.0%) and ethylenediamine (99.0%) were purchased from Alfa Aesar, Canada.

Titanium isopropoxide (98.0%) and hydrochloric acid (37.0%) were purchased from Acros

Organics, Canada. 2, 7- Dinitro-9-fluorenone was purchased from Tokyo chemical industry and

anhydrous ethanol (100%) was purchased from Commercial alcohols, Canada.

3.3.2 Synthesis of adsorbents

Adsorbent A, adsorbent B, and adsorbent C were prepared with the mesoporous Al2O3, Ti-

substituted mesoporous Al2O3 and commercial ϒ-alumina supports, respectively. There were three

steps involved in the syntheses of adsorbents. First, the supports, mesoporous Al2O3, Ti-substituted

mesoporous Al2O3 were synthesized, afterward, ethylenediamine (linker) was added onto the three

supports and as the third and final step, the immobilization of the π-acceptor onto the three

materials was performed.

Support

Linker

r

Π-acceptor

Figure 3.1 Schematic of synthesized adsorbent for desulfurization

29

The synthesis of mesoporous alumina and Ti-substituted mesoporous alumina followed the

method reported by Badoga et al. 2014 and Duprez and Wilson 2009 with a few modifications.

The synthesis procedures of adsorbents are described below.

Synthesis of Al2O3 and Ti-substituted mesoporous Al2O3 supports

For the synthesis of pristine alumina support, 39.5 g of aluminum isopropoxide was dissolved in

130 mL of ethanol contained in a 250 mL beaker placed in a hot plate with a mechanical stirrer for

1 h at 40°C (Solution A). Simultaneously, 260 mL of ethanol was placed in a 500 mL beaker at

40°C incorporated with a mechanical stirrer, and then 19.4 g of P-123 was added. The mixture was

stirred for 2 h at 40°C in order to dissolve P-123 (Solution B). Then, 29 mL of HNO3 was added

with an HNO3/H2O ratio of 0.2 (14mL of HNO3 and 70 mL of H2O). The next step was the addition

of the content of solution A to solution B dropwise and stirred for 2 h at 40°C. The resulting

material was aged for 4.5 days at 65°C in a Teflon bottle prior to calcination at 600°C in a muffle

furnace for 5 h with a ramp rate of 0.5°C/min.

For the synthesis of Ti-substituted mesoporous Al2O3 support, 43 g of aluminum isopropoxide and

3.5 g of titanium isopropoxide were dissolved in 117 mL of ethanol in a 500 mL beaker under

mechanical agitation for 30 min. Simultaneously, 17.9 g of P123 was dissolved in 240 mL of

ethanol at 40 °C for 30 min in another 500 mL beaker. Subsequently, 27 mL of HNO3 (70 wt.%)

was added to the beaker containing P123 solution and allowed to stir for another 15 min.

Afterward, the first mixture was added dropwise to the content of the second beaker under vigorous

stirring for 2 h. The mixture was then transferred into a Teflon bottle for aging at 65°C for 4.5

days. Finally, the solid material was recovered under vacuum filtration, dried at 80°C and calcined

at 600 °C for 5 h at a ramp rate of 0.5°C/min.

Amine functionalization on supports

The functionalization of the synthesized supports with ethylenediamine (EDA) provides anchoring

sites for the π-acceptor. The procedure for the immobilization of the amine linker on supports is

described below. Firstly, 12.1 g of EDA was added to a beaker containing 144 ml of methanol

under vigorous stirring until complete dissolution of the solid particles. Secondly, 10 g of the

support was added under continuous stirring for 3h. The resulting mixture was filtered, washed

30

with methanol and dried in a vacuum oven at 70 °C for 12 h. For comparison purposes, commercial

ϒ-alumina was also functionalized using the same linker.

Immobilization of π-acceptor on supports

87 mL of toluene and 6 mL of acetic acid were added to a 500 mL two-necked flask equipped with

a mechanical stirrer immersed in an oil bath at 100°C. Then, 1.06 g of DNF π-acceptor was added

under continuous stirring until completely dissolved. Subsequently, the functionalized support (i.e.

support + linker matrix) was added and stirred at 400 rpm under reflux at 100°C for 3 days. Once

the reaction was completed, the materials were recovered via vacuum filtration and washed with

toluene. Then it was dried under vacuum at 85 °C for 48 h.

3.3.3 Characterization of supports and adsorbents

To determine the textural properties of the synthesized supports, supports + linker as well as the

formulated support-linker-π acceptor adsorbents, the N2 adsorption-desorption BET technique was

used. The nitrogen physisorption was performed using a Micrometrics ASAP 2020 analyzer with

liquid nitrogen at a temperature of 77 K to determine the specific surface area, pore volume and

average pore diameter. The Brunauer-Emmett-Teller (BET) method was used in the pressure range

(P/Po) of 0.05-0.30. The adsorption-desorption sections of isotherms were used to calculate the

pore diameter and pore size of the synthesized materials using the Barrett-Joyner-Halenda (BJH)

method. Using the N2 amount adsorbed by the materials at a P/Po = 0.4, the volume of the

mesopore was calculated. Using the amount of N2 adsorbent at a P/Po=0.95, the total pore volume

was calculated with the assumption that the external surface was negligible as compared to the one

in the pores. In all cases, a correlation coefficient of 0.99 was obtained.

The Fourier transformed infrared spectrometer (FT-IR) was used to qualitatively determine the π-

acceptor presence on the materials. The spectra were recorded in the range of 400-4000 cm-1

wavenumbers, which used 128 scans with a nominal resolution of 4 cm-1 with a VERTEX 70 FT-

IR instrument equipped with a diamond ATR.

The thermal stability of the samples was determined using a TA instruments Q500 series

thermogravimetric analyzer (TGA). Nitrogen was used to purge the samples and then the analysis

was performed. Using 10-20 mg of sample weighed into a high-temperature platinum pan, the

31

sample was heated to 800°C at a ramp rate of 10°C/min. As the sample was being heated, the

weight loss of the samples was measured as a function of the increasing temperature.

The X-ray diffraction (XRD) patterns of the supports and synthesized adsorbents were obtained

using a Bruker D8 Advance X-ray diffractometer (Bruker AXS, Karlsruhe, Germany). The XRD

analysis was acquired with the application of monochromatic CuKα radiation (λ = 0.15418 nm)

generated at 40 kV voltage and 130 mA. The diffraction patterns were measured between the wide-

angle (2θ) of 10° and 90° at a scanning speed of 0.2°/min.

Around 4-6 mg of each sample were combusted using an El Vario III CHSN elemental analyzer

(Elemental Americas Inc., Mt. Laurel, NJ, USA) was used to determine the percent content (wt/wt)

of carbon (C), hydrogen (H), and nitrogen (N) atoms. It was also used to determine the nitrogen

concentration in between synthesis stages.

All X-ray Photoelectron Spectroscopy (XPS) measurements were collected using a Kratos

(Manchester, UK) AXIS Supra system at the Saskatchewan Structural Sciences Centre (SSSC),

the University of Saskatchewan. This system is equipped with a 500 mm Rowland circle

monochromatic Al K-α (1486.6 eV) source and combined hemi-spherical analyzer (HSA) and

spherical mirror analyzer (SMA). A spot size of hybrid slot (300x700 microns) was used. All

survey scan spectra were collected in the 5-1200 binding energy range in 1 eV steps with a pass

energy of 160 eV. High-resolution scans of multiple regions were also conducted using 0.05 eV

steps with a pass energy of 20 eV. An accelerating voltage of 15 keV and an emission current of

15 mA were used for the analysis.

3.3.4 Procedure for adsorption experiments

A model feed was prepared in order to test the adsorption efficiency towards sulfur compounds of

the synthesized materials in a batch reactor system. For the model feed, thiophene was dissolved

in ultra-low sulfur diesel (ULSD). The ULSD contains 27 and 12 ppm for nitrogen and sulfur,

respectively. For the desired concentration of 500 ppm, 0.0656 g of thiophene was diluted in 50

ml of ULSD, based on the formula shown below.

𝑝𝑝𝑚 (𝑆) = 𝑤𝑡. 𝑡ℎ𝑖𝑜𝑝ℎ𝑒𝑛𝑒 (𝑔) 𝑋 32.065

84.14 𝑋

1

50 𝑋 106 (3.1)

32

Then, the adsorption studies were carried out by adding adsorbent to model feed (1:5 ratio, by

weight). The feed was contacted with the adsorbent by placing 0.5 g of the adsorbent and 2.5 g of

the model feed in a glass vial with a cap. Using a magnetic stirrer at 400 rpm, the adsorbent and

the model feed were mixed at 22°C, for 12 hours. After mixing, the liquid product was separated

from the adsorbent by vacuum filtration and analyzed using an Antek N/S analyzer. The sulfur

removal efficiency of adsorbent was calculated based on the formula:

3.4 Results and discussion

3.4.1 N2-adsorption measurement

The isotherms, as well as pore size distributions for pristine supports and their corresponding

adsorbents, were determined by N2-adsorption/desorption measurement. Figure 3.2 shows the

isotherms of the three supports (mesoporous Al2O3, Ti-substituted mesoporous Al2O3 and

commercial γ- alumina) and DNF-immobilized supports (Al2O3-EDA-DNF and Ti-substituted

Al2O3-EDA-DNF and commercial γ- alumina-EDA-DNF). The type IV isotherm confirms the

presence of textural mesoporosity in the pristine supports and their respective adsorbents

(Satterfield 1996). Since adsorption predominantly proceeds on the surface, the porosity of the

adsorbent material plays a crucial role. This is because the porosity controls the mass transfer

processes via charge transfer complexation between the π-acceptor of the adsorbent material and

sulfur compounds (π donors) of the feed; thus, facilitating the adsorption of sulfur compounds and

their subsequent desulfurization.

S removal efficiency (%)=S content in untreated feed (ppm)-S content in treated feed (ppm)

S content in untreated feed (ppm) 𝖷 100 (3.2)

33

Figure 3.2 N2-adsorption/desorption isotherms for Al2O3,Ti-substituted meso-alumina and

commercial gamma-alumina their corresponding DNF-immobilized supports (Al2O3-EDA-DNF

and Ti-substituted meso-alumina -EDA-DNF).

The pore size distribution profiles of the prepared adsorbents are given in Figure 3.3. It can be

observed from the figure that the cumulative volume of N2 desorbed by the adsorbent A is more

than that of the adsorbent B. As given in Table 3.1, the incorporation of Ti in the framework of

Al2O3 significantly changed the textural properties of mesoporous alumina. It can be noticed that

the surface area and pore volume significantly decreased after the immobilization of the linker and

π-acceptor on all three supports.

0

200

400

600

800

1000

0 0.2 0.4 0.6 0.8 1

Vad

s (c

m³/

g S

TP

)

Relative Pressure (P/Po)

Mesoporous alumina support

Ti-substituted meso-alumina support

Commercial gamma-alumina support

Mesoporous alumina support-EDA-DNF

Ti-substituted meso-alumina- EDA-DNF

Commercial gamma-alumina-EDA-DNF

34

Figure 3.3 Pore size distribution of the DNF immobilized support Al2O3-EDA-DNF, Ti-

substituted Al2O3-EDA-DNF and commercial alumina-EDA-DNF.

The Al2O3–EDA and Ti-substituted Al2O3–EDA materials suffered a 23.9 and 26.7 % reduction

in surface areas, respectively; while a reduction of 28.5 and 27.3 %, respectively, was recorded in

their pore volumes. That notwithstanding, significant losses of both surface areas and porosities

were observed in the final adsorbents after the DNF π-acceptor was immobilized, whereas the

reduction in specific surface area for the final material was in the range of 80-85% and that of pore

volume was in the range of 84-95% for the Al2O3 and Ti-substituted meso- Al2O3 based absorbents.

The observed drastic changes in textural properties, typically, the pore volume, could be attributed

to the bulky nature of the DNF π-acceptor. The adsorbent C, which is based on γ- alumina did not

show significant changes in the textural properties like the adsorbents A and B.

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0 50 100 150

dV

/dlo

g

Pore width (nm)

Adsorbent A

Adsorbent B

Adsorbent C

35

Table 3.1 Textural properties of the synthesized support, support-linker, and support-linker-π-

acceptor.

Sample name BET surface

area (m2/g)

Pore volume

(cm3/g)

Pore diameter

(nm)

Supports

Alumina 548 1.37 6.7

Titania-alumina* 341 0.99 7.6

Commercial alumina 241 0.53 6.0

Supports + EDA

Alumina-ethylenediamine 417 0.98 6.3

Titania-alumina-

ethylenediamine

250 0.72 7.0

Commercial alumina-

ethylenediamine

264 0.63 6.2

Supports + EDA + DNF

Alumina-ethylenediamine-

dinitrofluorenone (A)

108 0.22 8.7

Titania-alumina-

ethylenediamine-

dinitrofluorenone (B)

66 0.07 5.0

Commercial alumina-

ethylenediamine-

dinitrofluorenone (C)

195 0.35 7.7

3.4.2 Fourier transform infrared analysis

Figure 3.4 displays the FTIR spectra of the alumina-based supports used, the broad bands from

400 to 1100 cm-1 are observed in all mixed oxides, it can be attributed to the superimposition of

the Al-O vibrations bands. The spectra of mixed oxides show a broad band between 2700-3750

cm-1 that corresponds to hydroxyl groups of oxides and surface adsorbed water (Badoga et al.,

2014). The strong peak at 1660 cm-1can be attributed to the stretching of –OH (Ahmed, 2011;

Duprez et al., 2011). The bands for Ti−O were obscured by the bands related to Al−O bonds,

hence, could not be identified (Silva et al., 2015). The π-acceptor, DNF contains two nitro groups

on fluorenone moiety. Figure 3.5 shows the adsorbents, ethylenediamine and dinitrofluorenone

spectra. The immobilization of the π-acceptor on all three supports is evidenced by the appearance

36

of nitro group (O=N–O) bands at 1340 cm-1 and 1570 cm-1 (Misra et al., 2015). The FTIR study

indicates that the desired adsorbents with the π-acceptor were synthesized successfully.

Figure 3.4 FT-IR spectra of the alumina-based supports

Figure 3.5 FT-IR spectra of adsorbents, ethylenediamine and dinitrofluorenone.

400 900 1400 1900 2400 2900

Tra

nsm

itte

nce

(a.u

)

Wavenumber (cm-1)

Adsorbent A

Adsorbent B

Adsorbent C

37

3.4.3 Thermogravimetric analysis

Thermal degradation profiles of the three adsorbents are shown in Figure 3.6. The prepared

adsorbents are stable up to 200 °C, a temperature that is within the working range of adsorption

reactions. The adsorbents showed significant weight loss after 280°C.

Figure 3.6 Thermal degradation profiles of adsorbents and dinitrofluorenone

As discussed earlier, the FTIR and XRD studies proved the presence of significant amount of

DNF in all three adsorbents. Prior to immobilization, supports were calcined at 600°C, and no

other compounds or impurities other than DNF is expected in this temperature region. TGA of

pure DNF was conducted to validate the above prediction. TGA of pure DNF also shows the

weight loss in the same temperature region as adsorbents (Figure 3.6). This result confirms that

the weight loss of adsorbents after 280°C is due to thermal degradation of DNF. As discussed

previously, the surface area and pore volume values drop as the concentration of DNF increases

in adsorbents. The TGA study showed ~20 % more DNF in adsorbent A than adsorbent C.

0

20

40

60

80

100

0 100 200 300 400 500 600

Wei

gh

t lo

ss (

a.u

)

Temperature (°C)

Adsorbent A

Adsorbent B

Adsorbent C

2,7-dinitro-9-

fluorenone (DNF)

38

Adsorbent A showed a significant drop in the surface area and pore volume values due to the

presence of the higher amount of DNF in it.

3.4.4 X-ray diffraction

X-ray diffraction of supports and adsorbents are shown in Figures 3.7 and 3.8. The XRD patterns

concluded that the mesoporous alumina and Ti-substituted mesoporous alumina supports are

mostly amorphous and the commercial alumina support contains the crystalline γ-alumina phase

(peaks at 46.3 ° and 67° ). The XRD patterns of the synthesized adsorbents (A, B and C) shown

in Figure 3.7 match to the π-acceptor, DNF, which is highly crystalline, therefore its peaks

dominant over the support XRD peaks.

Figure 3.7 X-ray diffraction for alumina supports. (A) Mesoporous alumina support, (B) Ti-

substituted meso-Al2O3 and (C) commercial γ-alumina supports.

10 30 50 70 90

Inte

nsi

ty (

a.u

)

2θ (degree)

A

B

C

39

Figure 3.8 X-ray diffraction for alumina based adsorbents and π-acceptor compound.

3.4.5 X-ray photoelectron spectroscopy

Adsorbent B was investigated by XPS to determine its Titanium coordination. In the mesoporous

materials, the framework and non-framework Ti species exist in the tetrahedral and octahedral

coordination, respectively. The XPS spectrum of adsorbent B shown in Figure 3.9 evidences the

presence of only tetrahedrally coordinated titanium by the appearance of characteristic bands at

458 and 462 eV. This values are in accordance with the other reports (Gonzalez-Elipe et al., 1989;

Biesinger et al., 2010; Lazzaroni and Hecq, 2003; Pouilleau et al., 1997).

10 30 50 70 90

inte

nsi

ty (

a.u

)

2θ(degree)

Adsorbent A

Adsorbent B

Adsorbent C

2,7 - dinitro-9-fluorenone

40

Figure 3.9 X-ray photoelectron spectroscopy of Adsorbent B.

3.4.6 CHNS elemental analysis

The results and their respective standard deviations are shown on Table 3.2. It can be observed

that when the linker (EDA) was attached to mesoporous alumina, the nitrogen composition

increases to 1.17%, this change in the nitrogen composition was expected as the linker (EDA) is

an amine, therefore its structure C₂H₄(NH₂)₂ containing nitrogen. Then, when the π-acceptor is

added we see that the nitrogen composition did not change significantly, this can be explained with

the synthesis procedure. Attaching the linker to the support does not involve a higher temperature

Ti2p

Name

Ti 2p

Pos.

456.10

FWHM

1.53

Area

153.79

At%

100.00

Ti 2

p

x 101

152

154

156

158

160

162

164

166

168

170

CP

S

468 464 460 456 452Binding Energy (eV)

41

other than 22°C, however, when the π-acceptor was synthesized, a temperature of 100°C was

needed for the reaction. With this being said, it might be possible that some of the linker was not

washed out after its synthesis onto the support and when the π-acceptor immobilization took place,

the linker in excess was washed off but the composition did not decreased, as the π-acceptor (DNF)

is high in nitrogen composition. The DNF chemical formula, C13H6N2O5, shows that is not only

rich in nitrogen, but also in carbon (C). This can explained why after immobilizing the π-acceptor

onto the supports and linkers, each sample shows an increase on its carbon composition. This

nitrogen composition behavior was not the case for the three supports, Ti-substituted meso-

alumina and commercial γ-alumina, where nitrogen compositions increased after adding the linker

and π-acceptor. However, the carbon composition behavior was the same in all of the materials.

Table 3.2 CHNS elemental composition of supports, support + linker and final adsorbent

Elemental composition % (wt/wt)

Sample name N (%) C (%) H (%) S (%)

Mesoporous alumina

support

0.00 ± 0.00 0.49 ± 0.00 3.44 ± 0.26 0.08 ± 0.11

Mesoporous alumina

support-EDA

1.17 ± 0.00 2.30 ± 0.05 3.04 ± 0.01 0.07 ± 0.02

Mesoporous alumina

support-EDA-DNF

1.17 ± 0.07 27.38 ±

0.36

3.77 ± 0.08 0.11 ± 0.03

Ti-substituted meso-

alumina

0.07 ± 0 0.77 ± 0.06 3.28 ± 0.97 0.02 ± 0.02

Ti-substituted meso-

alumina-EDA

0.70 ± 0.00 2.83 ± 0.81 3.40 ± 0.02 0.05 ± 0.01

Ti-substituted meso-

alumina-EDA- DNF

0.82 ± 0.02 21.63 ±

1.43

3.40 ± 0.59 0.05 ± 0.00

Commercial γ-alumina 0.00 ± 0.00 0.15 ± 0.17 3.60 ± 0.33 0.04 ± 0.00

Commercial γ-alumina

-EDA-DNF

1.82 ± 0.11 15.46 ±

0.75

2.90 ± 0.88 0.03 ± 0.01

42

3.4.7 Adsorption experiments using ultra-low sulfur diesel (ULSD) feedstock

Adsorption experiments were performed for a 24 h duration to measure the desulfurization abilities

of adsorbents using a model feed with 500 ppm sulfur in ULSD. As shown in Figure 3.10,

adsorbents A, B and C removed 85.7, 55.3 and 90.4 wt. % sulfur, respectively. Among the

adsorbents, adsorbents A and C appear to be suitable materials for desulfurization. These

adsorbents are based on alumina supports, whereas the adsorbent B is made of Ti substituted Al2O3

support. As compared with alumina supports, Ti substituted Al2O3 has inferior surface area and

pore volume. In the same way, adsorbent B, which is formulated with Ti substituted Al2O3, has

lower surface textural properties. The poor surface area and pore volume of adsorbent B are

accountable for its inferior sulfur adsorption tendency.

Figure 3.10 Comparison of sulfur removal efficiency of Adsorbents A, B and C. Adsorption

parameters: Feed = 500 ppm of sulfur in ULSD, T = 22 °C, stirring = 400 rpm, atmospheric

pressure, adsorbent to feed ratio = 1:5, time = 24 h).

All adsorbents contain ethylenediamine and DNF on them. The possibilities of leaching of these

compounds were checked by conducting adsorption experiments using ULSD non-spiked feed

85.7

55.3

90.4

0

20

40

60

80

100

1

Su

lfu

r R

emo

va

l (w

t. %

)

Adsorbent AAdsorbent B Adsorbent C

43

with 27 and 12 ppm for nitrogen and sulfur, respectively. As shown in Figure 3.11, all three

adsorbents removed completely the total sulfur content, however, this was not the case for

nitrogen. The nitrogen contents of liquid products of adsorbents A and C are lower than that of the

starting feed. On the contrary, a slight increase in the nitrogen concentration is observed with the

adsorbent B. The linker and π-acceptor are nitrogen-based compounds. The leaching of these

compounds during adsorption can lead to an increase in the nitrogen concentration of the liquid

product. The increase of nitrogen concentration with the adsorbent B in Figure 3.11 evidences the

possibility of some leaching of either the linker or the π-acceptor compounds from the adsorbent

B.

Figure 3.11 Sulfur and nitrogen removal efficiency of adsorbents A, B and C with ULSD non-

spiked feed. Adsorption parameters: T = 22 °C, stirring = 400 rpm, atmospheric pressure,

adsorbent to feed ratio = 1:5, time = 24 h.

16.6

100

-3.6

100

21.7

100

-20

0

20

40

60

80

100

Rem

oval

(wt.

%)

Adsorbent A Adsorbent B Adsorbent C

Nitrogen

removalSulfur

removal

44

3.4.8 Effects of adsorption time

The time it takes by an adsorbent to remove sulfur and nitrogen impurities is important as it can

impact the adsorption capacity. Therefore, the effects of contact time of the adsorption process can

be considered as an important factor to determine whether a material is suitable for the adsorption

application or not. The effect of time on adsorption of sulfur species was studied by performing

the experiments for 1, 6, 12, 24, 48, 72 and 96 h by using the ULSD spiked feed with 500 ppm of

sulfur. As shown in Figure 3.12, the adsorption capacity is found to increase with the contact time.

During the first hour of the experiment, around 60% of sulfur impurities were removed by

adsorbent C and around 20% with adsorbents A and B. Then in the second sample, which was

taken after 6 h, there is an increase in the sulfur removal with all three adsorbents.

Figure 3.12 Effects of adsorption time on sulfur removal (Adsorption parameters: T = 22 °C,

stirring = 400 rpm, atmospheric pressure, adsorbent to feed ratio = 1:5, time = 0 to 96 h).

Adsorbent A removed 60%, adsorbent B removed around 40% and adsorbent C removed over

90%. The consecutive time samples for adsorbents B and C seem to stay within the same range,

however, the adsorption activity of the adsorbent A gradually increases and reaches its maximum

value at the contact time of 72 h. Though absorbents A and C showed similar desulfurization

activities after 72 h, the absorbent C is most preferred for the adsorptive desulfurization process,

45

as it needs the shorter equilibrium time. The most suitable adsorbent for industrial applications is

the one that possess a shortest contact time (shortest equilibrium time). As the contact time and

flow rate of feedstock are indirectly proportional, high flow rate operation can provide a cost-

effective adsorptive desulfurization process.

3.4.9 Effects of feed concentration on adsorption capacity

Adsorption isotherms are the equilibrium relationships that provide insights into the distribution

of adsorbate between two phases for a particular adsorbate and adsorbent system (Foo and

Hameed, 2010). Studying the equilibrium capacity of an adsorbent is helpful as it can help to

determine the maximum adsorption capacity of adsorbent in relation to adsorbate (Mckav and

Ireland, 1985). This adsorption equilibrium capacity is achieved when the rate at which the

molecules adsorb into the adsorbent is equal to the rate at which they desorb from it; this is

considered a dynamic concept. The concentration of the solute should not change in equilibrium

either in the solid or in the bulk solution, as the equilibrium is considered a characteristic of the

entire system (El Qada et al., 2006).

Here, sulfur model feeds with varying initial concentrations of thiophene in ULSD were used for

determining the adsorption capacity data. The sulfur adsorption capacity, q (mg/g) of the

adsorbents was calculated according to the following equation:

𝑞 = 𝑀𝑓 (𝐶𝑖−𝐶𝑒)

𝑀𝑠 (3.3)

Where, Mf is the model feed weight (g), Ci and Ce are the initial and final concentrations of the

sulfur in the model feed (ppm), respectively and Ms is the mass of solid adsorbent (g).

There are two types of adsorption; physical and chemical. Chemical adsorption constitutes creating

bonds in between adsorbate (sulfur compounds) and adsorbent, whereas physical adsorption is

based on weak intermolecular bonds forces like Van der Waals and hydrogen bonding (Nathan

and Scobell, 2012). The adsorbate can adsorbed on any sites of the adsorbent and in two different

ways: monolayer or multilayer. The monolayer adsorption constitutes one molecule getting

attached onto the surface of the adsorbent and multilayer is when several molecules may stack

onto the surface of the adsorbent (Hendricks, 2011).

46

As shown in Figure 3.13, the relation between the various initial feed concentrations compared to

their adsorption capacity seems to show the curve, which increases quickly in the beginning, but

the adsorption gain decreases and becomes more difficult as the sulfur concentration increases.

This adsorption pattern indicates that the adsorption surface of adsorbents A, B and C were

heterogeneous, resulting in both monolayer and multilayer adsorptions. Similar results of increase

in the adsorption capacity with an increasing amount of the initial adsorbent concentration were

reported (Fei et al., 2017; Ha et al., 2018). Increase of adsorbate concentration increases the

probability of collision between adsorbate and adsorbent, also the driving force of the

concentration gradient and thus causes an increase in adsorption capacity.

Figure 3.13 Effects of initial concentration of sulfur model feed on adsorption capacity of

adsorbents A, B and C (Adsorption parameters: T = 22 °C, stirring = 400 rpm, atmospheric

pressure, adsorbent to feed ratio = 1:5, time = 24 h).

3.5 Conclusions

Three different charge-transfer complex (CTC) adsorbents for desulfurization of liquid fuels were

prepared and characterized. The BET-N2 study evidenced the mesoporous nature of all three

adsorbents. The FTIR, TGA and XRD studies confirmed the successful incorporation of the π-

acceptor on the support. Adsorption experiments were conducted with a model feed of 500 ppm

sulfur. The CTC adsorbents prepared based on mesoporous alumina (Adsorbent A) and

commercial ϒ-alumina (Adsorbent C) showed higher sulfur removal than the Ti incorporated

material (Adsorbent B). The poor performance of adsorbent B is due to leaching of the linker and

47

π-acceptor in the adsorption medium (ULSD). No leaching of the linker and π-acceptor was

observed with adsorbents A and C. The adsorption capacity profiles of adsorbents A and C are

logarithmic with the monolayer and multilayer adsorptions. Among adsorbent A and C, it is

concluded that adsorbent C is more suitable for sulfur removal due to the shorter equilibrium time

of adsorption.

48

Chapter 4 : Optimization, regeneration and kinetic studies

This chapter presents the experimental results and discussions of the third phase of this research

work. Section 4.1 details the chemical used in this part of the research along with those used for

the synthesized adsorbents; it also briefly describes the preparation of the adsorbents previously

detailed in Chapter 3. In section 4.2, the optimization design of the experiments is presented and

the results are further discussed. Kinetic studies are discussed in section 4.3, two kinetic models

were used to fit the data and further explain the adsorptive behavior of the adsorbents.

Thermodynamic studies are presented in section 4.4 where free activation energy, enthalpy,

entropy and activation energy are calculated and provide an insight of the thermodynamic side of

the adsorptive behavior of these alumina-based adsorbents.

4.1. Experimental methods

4.1.1 Materials/ Chemicals

Poly (ethylene oxide)- block poly (propylene oxide) - block poly (ethylene oxide) (P123), carbon

tetrachloride (CCl4) and dichloromethane were purchased from Sigma Aldrich, Ontario, Canada.

Nitric acid (≥70.0%), methanol (>99.0%), toluene (>99.0%) and acetic acid (>99.0%) were

purchased from Fisher Chemical, Saskatoon, Canada. Aluminum isopropoxide (98.0%) and

ethylenediamine (99.0%) were purchased from Alfa Aesar, Saskatoon, Canada. Titanium

isopropoxide (98.0%) and hydrochloric acid (37.0%) were purchased from Acros Organics,

Saskatoon, Canada. 2, 7- Dinitro-9-fluorenone was purchased from Tokyo chemical industry and

anhydrous ethanol (100%) was purchased from Commercial alcohols, Saskatoon, Canada.

4.1.2 Preparation of adsorbents

The synthesis method of the charge transfer adsorbents used in this study was described in detail

in our previous publication (Botana de la Cruz et al., 2020). In brief, adsorbents A, B, and C were

prepared with the mesoporous Al2O3, Ti-substituted mesoporous Al2O3 and commercial ϒ-

alumina supports, respectively. The syntheses of adsorbents involved (i) preparation of supports,

(ii) anchoring of linker, ethylenediamine on supports, and (iii) immobilization of a π-acceptor, 2,7-

dinitro-9-fluorenone on supports.

49

4.2 Process parameter Optimization

To investigate the influence of process parameters and their combined interactions on the

adsorptive desulfurization process, a set of experiments was designed by varying process

parameters, namely, temperature, adsorption time and adsorbent loading. The model feed

concentration was kept constant. Experiments were designed by the Central Composite Design

(CCD) approach using the Design Expert® software (version 6.0.11, State-Ease Inc., Minneapolis,

USA). Briefly, the CCD method is use to optimize the conditions of a known multi-variable system

as it evaluates the influence of the independent variables in a set of experiments given (Vaez et al.,

2012).

Table 4.1 Optimization parameters and their corresponding range used for the Central

Composite Design.

Variables Symbols Range

Temperature (˚C) X 22 60

Time (min) Y 15 60

Adsorbent loading (g) Z 0.25 0.75

Table 4.1 shows the ranges of variables used to design. As given in Table 4.2, a set of 19

experiments (14 unique) was designed, with the center point experiment replicated 5 times.

Experiments were repeated to check reproducibility under the same set of conditions. The model

feed used in this study contained 541 ppm of sulfur which was prepared by dissolving thiophene

in ultra-low sulfur diesel (ULSD). The typical adsorption experiment was carried out placing 0.5

g of the adsorbent and 2.5 g of the model feed in a glass vial with a cap. Using a magnetic stirrer

at 400 rpm, the adsorbent and the model feed were mixed at 22˚C for 24 h and the liquid product

was separated from the adsorbent by vacuum filtration and analyzed using an Antek N/S analyzer.

50

Table 4.2 Design of experiments for adsorption of sulfur and nitrogen species

4.2.1 Adsorption parameters optimization studies

The effects of time, temperature and adsorbent loading on sulfur removal were determined by

plotting a three-dimensional (3D) response surface and control graphs (see Figure 4.1). Adsorbent

A shows its highest adsorption activity with 76.3% of sulfur removal at an adsorbent loading of

20%, 41°C and a reaction time of 43 min. However, at a temperature of 22°C, an adsorbent loading

of 30% and a reaction time of 27 min, adsorbent A shows 74.0% sulfur removal. In terms of

energy requirements, adsorption at 41°C might not be ideal as compared to 22°C.

Besides, raising the temperature by 20˚C increased the sulfur removal only by 2 wt.%. Adsorbent

B showed a maximum sulfur removal of 57.2 wt.% at an adsorbent loading of 10% (0.25 g), a

temperature of 60°C and an adsorption time of 27 min. Similarly, adsorbent C reached its

maximum adsorption capacity at an adsorbent loading of 30% (0.75 g), a temperature of 22°C and

Run

number

Adsorbent

loading (g)

Temperature

(°C)

Time

(min)

1 0.75

22

60

2 0.75 27

3 0.25 60

4 0.25 27

5 0.75

60

60

6 0.75 27

7 0.25 60

8 0.25 27

9 0.5

41

43

10 0.5 43

11 0.5 71

12 0.5 43

13 0.5 43

14 0.5 43

15 0.5 15

16 0.5 43

17 0.92 43

18 0.08 43

19 0.5 73 43

51

an adsorption time of 27 min, resulting a sulfur removal of 88.7 wt. %.

Figure 4.1 The three-dimensional response surfaces: effects of temperature and adsorbent

loading on desulfurization activity of adsorbents A, B and C.

To fit the response function on sulfur removal, regression analysis was performed. The model

equation for sulfur removal for adsorbents A, B and C are shown in Table 3. The individual effect

of each of the process parameters such as temperature (X), time (Y) and adsorbent loading (Z),

and combined effects of two of the process parameters on sulfur adsorption can be distinguished

by using equations given in Table 3. For example, the effects of X-Y, X-Z, and Y-Z are not as

significant as the individual effect of X, Y and Z on the sulfur adsorption capacity of the adsorbent

A.

52

Table 4.3 Linear equations for adsorbents A, B and C from Composite Design Expert based on

the proposed set of experiments.

Adsorbent Linear equation

A Total S removal= -5.37+0.17X + 111.04Y +68.99Z - 0.60XY - 0.44XZ -

75.31YZ

B Total S removal= 50.73 + 0.34X – 55.31Y – 19.13Z + 0.30XY – 0.17XZ +

48.46 YZ

C Total S removal = 103.46 – 0.01X – 1.63Y – 104.90Z – 0.54XY + 0.27XZ +

116.77YZ

Note: Where X,Y and Z are temp.(˚C), time (h) and adsorbent loading (wt.%), respectively.

4.3 Adsorption kinetics

Apart from the results of the experimental studies, mathematical modeling is an efficient method

for analysis of any chemical process. Theoretical and empirical models are used to fit experimental

data. Using this method provides information to understand and simulate the adsorptive removal

of sulfur in from fuels that can potentially reduce the cost of experiments and increase the

feasibility of the adsorption process. In this case, empirical models were used, pseudo-first and

pseudo-second order kinetic model; these models are known for their simplicity when modelling

experimental data.

Kinetic studies were conducted in a stirred batch system consisting of a 15 mL glass vial in which

a mixture of the adsorbent and the model feed was stirred at 400 rpm. The adsorbent to feed ratio

was maintained at 1:5. To determine the minimum time required for the adsorbent to reach

adsorption equilibrium, the experiments were carried at different times ranging from 15 min to 72

h. To investigate the adsorption mechanism, the temperature was kept constant at 22°C.

After the duration for adsorption was elapsed, the adsorbent was filtered under vacuum and the

treated feed was collected in a vial for total sulfur and nitrogen analysis using an Antek analyzer.

Following the N/S analysis, the adsorption efficiency was calculated using the formula shown

below:

53

𝑆 𝑟𝑒𝑚𝑜𝑣𝑎𝑙 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 (%) = 𝑆 𝑖𝑛 𝑡ℎ𝑒 𝑢𝑛𝑡𝑟𝑒𝑎𝑡𝑒𝑑 𝑓𝑒𝑒𝑑 (𝑝𝑝𝑚)−𝑆 𝑖𝑛 𝑡ℎ𝑒 𝑡𝑟𝑒𝑎𝑡𝑒𝑑 𝑓𝑒𝑒𝑑

𝑆 𝑖𝑛 𝑡ℎ𝑒 𝑢𝑛𝑡𝑟𝑒𝑎𝑡𝑒𝑑 𝑓𝑒𝑒𝑑 (𝑝𝑝𝑚)× 100 (4.1)

The equilibrium capacity of the adsorbents was calculated using the following formula:

𝑞𝑒 = 𝐶0−𝐶𝑒

𝑚𝑉 (4.2)

Where C0 and Ce are the initial and equilibrium concentrations of sulfur (mg L-1) respectively, V

is the volume of the feed used (L), and m is the mass of the adsorbent (g).

4.3.1 Kinetic studies

To analyze the data from the adsorption experiments, two conventional kinetic models were used,

pseudo-first and pseudo-second order models (Boahene et al., 2013; Malik, 2004). Figure 4.2

shows the adsorption kinetic curves for the pseudo-first-order model of adsorbents A, B and C for

sulfur removal. The pseudo-first order model formula was rearranged for linearized data plotting

and is shown below:

ln(𝑞𝑒 − 𝑞𝑡) = ln 𝑞𝑒 − 𝑘1 ∗ 𝑡 (4.3)

Where qe and qt are amounts of sulfur compound adsorbed at equilibrium (qe) and at various times

(qt) calculated in (mg g-1) and k1 is the rate constant of the pseudo-first-order model of adsorption

(min-1). The values of qe and k1 can be calculated using the intercept and slope of the linear plot of

ln (qe - qt) versus t (see equation 4.3).

(4.4)

𝑡

𝑞𝑡=

1

𝑘2𝑞𝑒2

+𝑡

𝑞𝑒

54

Figure 4.2 Pseudo first-order adsorption kinetics of adsorbents A, B and C. Temperature =22°C,

feed to adsorbent ratio=5. The solid line is the results for the kinetic model; the dash line is

pseudo first-order-model simulation

For the pseudo-first order model, values of qe and k1 were calculated, which are shown in Table

4.4, along with other kinetic parameters and correlation coefficients. The experimental data

deviated considerably from the theoretical data. Also, the theoretical qe values found from the first-

order kinetic model did not give reasonable values. The kinetic parameters of the pseudo-second-

order model are also presented in Table 4.4. From this table, it can be seen that the pseudo-second

order model (see equation 4.4) adequately fits the adsorption experimental data for all three

adsorbents. This observation was corroborated by the correlation coefficient R2 values obtained

from this model. For adsorbents A and C the R2 values were higher than for adsorbent B; however,

the pseudo-first-order model did not show R2 higher than 0.5. For the pseudo-second order model,

both adsorbents A and C showed R2 values greater than 0.99; suggesting a better correlation of the

experimental data to the kinetic model. As it can be observed on Figure 4.3

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

0 1000 2000 3000 4000 5000

qt(

mg/g

)

time (min)

Adsorbent AAdsorbent BAdsorbent C

55

Figure 4.3. Pseudo first-order adsorption kinetics of adsorbents A, B and C. Temperature =22°C,

feed to adsorbent ratio=5. The solid line is the results for the kinetic model; the dash line is

pseudo first-order-model simulation.

Table 4.4 Kinetic parameters for pseudo-first and pseudo-second-order model

Pseudo-first order model Pseudo-second order

model

Adsorbent qe (exp)

(mg·g-1) qe(cal) (mg·g-1) K1(min-1) R2

qe (cal)

(mg·g-1)

K2(g·mg-

1·min-1) R2

A 2.79 -0.00112 3.5E-04 0.39 1.01 1.59E-02 0.99

B 2.08 -0.00021 3.4E-05 0.18 1.20 1.22E-02 0.92

C 2.78 -0.00117 -9.0E-04 0.33 1.03 9.21E-03 0.99

0

200

400

600

800

1000

1200

1400

1600

1800

0 1000 2000 3000 4000 5000

t/qt

time (min)

Adsorbent AAdsorbent BAdsorbent C

56

𝐾𝐷 =𝑞𝑒

𝐶𝑒

Values for qe (exp) and qe (cal) highly differ for the pseudo-first and second order models, the

negative k1 values indicates that they are not significant for the model (Pan and Xing, 2010), as

can be expected when looking at the R2 values. Even though the qe (exp) value for adsorbent B

seems to be closer to the qe (cal); this does not mean that it depends entirely on the adsorptive

behavior. The R2 for the pseudo-second order model for adsorbent B is the lowest with a value of

~0.92 when compared to those for adsorbents A and C; meaning that the pseudo-second order

model fits better for adsorbents A and C adsorption data. The pseudo-second order model is also

known as the double-exponential model or a two-step mechanism (Li et al., 1999). Michard et al.

1996, proposed a diffusion control mechanism consisted of two-steps. In the first one, a rapid

uptake of adsorbate takes place; this involves internal and external diffusion. In the second step,

the adsorption rate is controlled by the intraparticle diffusion before the adsorption reaches

equilibrium. Therefore; the pseudo-second order model can be consider a diffusion model.

4.4 Adsorption thermodynamics

Using the reaction rate constant of the pseudo-second-order model, the activation energies of the

three adsorbents were calculated following a method reported by Laidler, 1984. Using the

Arrhenius equation, the activation energy, E2, was calculated as shown below:

ln 𝐸2 = −𝐸𝑎

𝑅𝑇+ ln 𝐴 (4.5)

Moreover, using the data obtained from the adsorption experiments, the free activation energy

(ΔG0) was determined by assuming that the adsorption process happened spontaneously. The

changes in enthalpy (ΔH0) and entropy (ΔS0) were also calculated using the equations shown

below:

(4.6)

(4.7)

ln 𝐾𝐷 =∆𝑆0

𝑅−

∆𝐻0

𝑅𝑇 (4.8)

∆𝐺0 = −𝑅𝑇 ln 𝐾𝐷

57

Where ΔG0 (kJ·mol-1) was calculated using the Van´t Hoff equation, R is the universal gas

constant, 8.314 J·mol-1, T is the absolute adsorption temperature in K, KD is the adsorption

equilibrium constant. Using the slope and intercept of the plot, ln KD vs. 1/T, ΔH0 and ΔS0 were

calculated.

4.4.1. Adsorption thermodynamic studies

As part of this study, the activation energy (Ea) was calculated to provide an insight into the

adsorption mechanisms. Using the Arrhenius equation, the activation energies were calculated for

adsorbents A, B and C. Using temperature in the range of 295-314 K, the activation energy values

of 8.27, -7.62 and 8.29 kJ·mol-1 were calculated for adsorbents A, B and C and are shown in Table

4.5. The values were obtained using the intercept and slope from the plot of ln k2 vs. 1/T for the

three adsorbents (see Appendix A).

Table 4.5 Activation energy calculated values for adsorbents A, B and C.

Adsorbent Activation energy (Ea)

(kJ/mol)

A 8.27

B -7.62

C 8.29

An activation energy Ea of 5~40 kJ·mol-1 corresponds to physical adsorption whereas 40~800

kJ·mol-1 corresponds to a chemical adsorption (Fei et al., 2017). The activation energy suggests

that adsorbents A and C remove sulfur compounds through physical adsorption. This means that

no chemical reactions take place during the adsorption, instead the sulfur molecules are attracted

to the adsorbent by weak forces like Van der Waals or hydrogen bonding (Duan et al., 2014). On

the other hand, adsorbent B has proven to have a poor adsorption activity. This can attributed to

its poor textural properties. Therefore, the activation energy values for adsorbent B can be

neglected.

The free activation energy (ΔG0), enthalpy (ΔH0) and entropy (ΔS0) are usually involved in an

adsorption process. To evaluate if the adsorption of an adsorbate can happen spontaneously or not,

the above parameters have to be determined. In this study, the free activation energy (ΔG0) values

58

for a temperature of 295 K were found to be 9.6, 13.0 and 7.5 kJ·mol-1, as shown in Table 4.6 for

adsorbents A, B and C, respectively. For a temperature of 314 K, the values were 10.6, 11.8 and

8.4 kJ·mol-1. The positive values of ΔG0 mean that adsorption of sulfur compounds on adsorbents

A,B and C is a slow process (Zhang et al., 2017). Using ln KD vs. 1/T values (see figures shown

in appendix B), enthalpy and entropy values were calculated for the three adsorbents using the

intercept and slope.

Table 4.6 Free activation energy values for adsorbents A, B and C.

ΔG (kJ/mol)

Temperature

(K)

Adsorbent

A B C

298 9.6 13.0 7.45

314 10.6 11.82 8.37

Enthalpy (ΔH0) values for adsorbents A, B and C were calculated to be 15.5, -19.2 and 15.3 kJ·mol-

1, respectively as shown in Table 4.7. Positive values of ΔH0 indicates that the adsorption process

is endothermic (Foo and Hameed, 2010), meaning that the reaction consumes energy. The change

of entropy (ΔS0) values in the adsorption system were -85.2, 20.9 and -77.1 J·mol-1 K-1 for

adsorbents A, B and C, respectively. Negative entropy values among adsorbents A and C indicate

the absence of significant change in the internal structure of the adsorbent during the adsorption

process. This is in accordance with the better textural properties of adsorbents A and C, compared

to adsorbent B as reported previously (Botana de la Cruz et al., 2020).

Table 4.7 Enthalpy and entropy values for adsorbents A, B and C.

Adsorbent ΔH (kJ/mol) ΔS (J/mol K)

A 7.91 -59.65

B -12.58 0.51

C 7.87 -52.19

59

4.5 Regeneration and reusability studies

Regeneration is a process in which the adsorbent is returned to its state prior to be used for

adsorption. This has to be easily performed in order to reuse the adsorbents without affecting the

viability of the process (Li et al., 2016). Thermal treatment, ultrasound and solvent extraction

method have been used to regenerate the spent adsorbent. Solvent extraction has been reported to

be the most efficient method to remove sulfur and nitrogen compounds (Han et al., 2014).

Considering there are no SOx and NOx emissions, regeneration using solvent extraction might be

consider an environmentally friendly process. Regeneration of the adsorbent is usually performed

so the adsorbent can undergo further regeneration. Reusability studies provide the necessary

information for industrial application as the adsorption process can be scale (Misra et al., 2018;

Zolotareva et al., 2019).

The reusability of adsorbents was studied at the same process conditions (1:5 adsorbent to feed

ratio, time=24 h, temperature= 22°C and 400 rpm) that were used for the screening of fresh

adsorbents. After the 24 h adsorption reaction, the adsorbent was regenerated using 3 different

solvents, chloroform, dichloromethane and carbon tetrachloride using a Soxhlet extraction set-up

shown in Figure 4.4.

Figure 4.4 Schematic of Soxhlet extraction set-up used for regeneration.

60

The spent adsorbent was placed inside the soxhlet apparatus on a thimble and extracted with 150

mL of solvent. During the Soxhlet extraction, the solvent was placed in a 250 mL round-bottom

flask connected to a reflux condenser in an immersed heated oil bath maintained at the boiling

point of each solvent. The extraction was carried out for 8 h, and then adsorbent was dried and

tested for reusability.

4.5.1 Regeneration and reusability studies

After the regeneration experiments, solvents were concentrated using a rotary evaporator and

analyzed for their sulfur levels. The sulfur extraction efficiencies of chloroform, carbon

tetrachloride and dichloromethane with adsorbent A (spent material) are shown in Figure 4.5.

Figure 4.5 Sulfur removal from Adsorbent A after 8 h of regeneration with chloroform, carbon

tetrachloride and dichloromethane

The highest sulfur removal was achieved using dichloromethane. It removed 80% of sulfur

molecules that were adsorbed on adsorbent A. The reusability of adsorbent A was studied for three

times by regenerating with DCM.

0.00

25.00

50.00

75.00

100.00

Chloroform CCl4 DCM

Sulf

ur

rem

oval

(w

t.%

)

Solvent

Chloroform

CCl4

DCM

61

Figure 4.6 Reusability results for adsorbent A using dichloromethane as solvent

The fresh adsorbent showed 96.5% of sulfur removal. After the first regeneration, the sulfur

removal efficiency dropped by only 7%. A significant drop was noted when adsorbent A was

reused for the third time. The nitrogen analysis of the solvent extract was carried out to determine

leaching of the immobilized EDA linker and π-acceptor during regeneration with DCM. The

solvent extract of first and second regeneration contained 86.0 and 93.5 ppm of nitrogen

respectively, due to leaching of EDA and DNF. The leaching of DNF from adsorbent A is expected

to drop the number of available adsorption sites and thus its adsorption capacity after successive

regenerations.

0

20

40

60

80

100

1st. 24 h adsorption 2nd. 24 h adsorption 3rd. 24 h adsorption

Rem

oval

(w

t.%

)

Sulfur removal (wt.%) Nitrogen removal (wt.%)

62

Chapter 5 : Conclusions and recommendations

5.1 Conclusions

The overall objective of this research work was to study the performance of π-acceptor

immobilized alumina-based materials as adsorbent for the removal of sulfur and nitrogen

impurities from liquid fuels. The goal was to improve the efficiency of hydrotreating in terms of

less catalyst loading, lower reaction temperature, and low severity conditions by using an

adsorption process as compared to typical hydrotreating process. The sub-objectives were (1) to

investigate the effects of the proposed synthesis method and its impact on the adsorbent properties

for sulfur and nitrogen compounds present in liquid fuels, and (2) to study process optimizations,

kinetics and long-term stability of potential adsorbent identified in above sub-objective.

Three different charge-transfer complex (CTC) adsorbents for desulfurization of liquid fuels were

prepared and characterized. The BET surface area measurement study evidenced the mesoporous

nature of all three adsorbents based on their textural properties; surface area, pore size and pore

volume. The FTIR, TGA and XRD studies confirmed the successful incorporation of the π-

acceptor on the support. Furthermore, adsorption experiments were conducted with a model feed

of 500 ppm sulfur. The CTC adsorbents prepared based on mesoporous alumina (Adsorbent A)

and commercial ϒ-alumina (Adsorbent C) showed higher sulfur removal than the Ti incorporated

material (Adsorbent B). The adsorbents successfully desulfurized the model feed used.

The poor performance of adsorbent B was due to leaching of the linker and π-acceptor in the

adsorption medium (ULSD). This does not support the hypothesis that TiO2 supported systems

show high activity. No leaching of the linker and π-acceptor was observed with adsorbents A and

C. The adsorption capacity profiles of adsorbents A and C are logarithmic with the monolayer and

multilayer adsorptions. Between adsorbent A and C, it is concluded that adsorbent C is more

suitable for sulfur removal due to the shorter equilibrium time of adsorption.

Furthermore, the optimization studies showed that the adsorption experiments should be conducted

under ambient conditions (~22˚C) as high sulfur removal was observed at this temperature. The

pseudo-second-order model had the best fit with the kinetic data of all three adsorbents. From the

kinetic studies, the activation energy (Ea) values were calculated, which indicated that the

adsorption is physical in nature when adsorbents A and C are used for adsorption. The values of

63

free activation energy suggested that the adsorption of sulfur compounds was a slow process.

Enthalpy values showed that adsorption on adsorbent A and C is endothermic. Entropy values

indicated that adsorbent A and C experienced no significant change in their internal structure. The

results of regeneration experiments showed that dichloromethane was relatively more effective as

a solvent for regeneration, and the adsorbent A can be reused at least three times.

5.2 Recommendations

Textural properties such as surface area, pore size and pore volume of the synthesized

supports can be improved and enhance the adsorptive desulfurization and denitrogenation

activity.

An increase of the nitrogen concentration of the batch adsorption experiments might

suggest leaching of the N-compounds (linker and π-acceptor) present in the adsorbent. The

effect of the linker and π-acceptor loading onto the support can be tested to avoid leaching

of these compounds onto the feed.

For this study, pyridine and thiophene were used as nitrogen and sulfur compounds in the

model feed. However, the presence of more nitrogen and sulfur compounds might show a

different adsorptive efficiency among the adsorbents. Introducing more nitrogen and sulfur

compounds on the feed can be explored and the affinity towards certain compounds can be

analyzed.

The used solvent should be separated from the pollutants and recycled. Also, the separated

sulfur and nitrogen compounds should be reutilized for improving the economics of the

entire process. Finally, techno-economic analysis of the entire process should be carried

out for determining the viability of this technology.

64

References

A.R Gonzalez-Elipe, G. Munuera, J. .. Espinos. 1989. “Compositional Changes Induced by 3.5

KeV Ar+ Ion Bombardment in Ni-Ti Oxide Systems.” Surface Science 220:368–80.

Abedi, Ali, Jackson Chitanda, Ajay K. Dalai, and John Adjaye. 2015. “Synthesis and Application

of Functionalized Polymers for the Removal of Nitrogen and Sulfur Species from Gas Oil.”

Fuel Processing Technology 131:473–82.

Abedi, Ali, Jackson M. Chitanda, Ajay K. Dalai, and John Adjaye. 2016. “Graft

Copolymerization of Glycidyl Methacrylate and Ethylene Glycol Dimethacrylate on

Alumina for the Removal of Nitrogen and Sulfur Compounds from Gas Oil.” Industrial and

Engineering Chemistry Research 55(35):9408–15.

Ahmed, M. A. 2011. “Structural and Nano-Composite Features of TiO 2 – Al 2 O 3 Powders

Prepared by Sol – Gel Method.” 509:2154–59.

Akelah, Ahmed and David C. Sherrington. 1981. “Application of Functionalized Polymers in

Organic Synthesis.” Chemical Reviews 81(6):557–87.

Almarri, Masoud, Xiaoliang Ma, and Chunshan Song. 2009. “Selective Adsorption for Removal

of Nitrogen Compounds from Liquid Hydrocarbon Streams over Carbon- and Alumina-

Based Adsorbents.” Industrial and Engineering Chemistry Research 48(2):951–60.

Alphonse, Pierre and Benjamin Faure. 2013a. “Microporous and Mesoporous Materials

Synthesis of Highly Porous Alumina-Based Materials.” Microporous and Mesoporous

Materials 181:23–28.

Alphonse, Pierre and Benjamin Faure. 2013b. “Synthesis of Highly Porous Alumina-Based

Materials.” Microporous and Mesoporous Materials 181:23–28.

Auerbach, Scott, Kathleen Carrado, Prabir Dutta, Shivaji Sircar, and Alan Myers. 2003. “Gas

Separation by Zeolites.” Handbook of Zeolite Science and Technology.

Badoga, Sandeep, Rajesh V. Sharma, Ajay K. Dalai, and John Adjaye. 2014. “Hydrotreating of

Heavy Gas Oil on Mesoporous Mixed Metal Oxides (M-Al2O3, M = TiO2, ZrO2, SnO2)

Supported NiMo Catalysts: Influence of Surface Acidity.” Industrial and Engineering

Chemistry Research 53(49):18729–39.

Badoga, Sandeep, Rajesh V. Sharma, Ajay K. Dalai, and John Adjaye. 2015. “Synthesis and

Characterization of Mesoporous Aluminas with Different Pore Sizes: Application in NiMo

Supported Catalyst for Hydrotreating of Heavy Gas Oil.” Applied Catalysis A: General

489:86–97.

Badoga, Sandeep, Ketaki Sohani, Ying Zheng, and Ajay K. Dalai. 2017. “Mesoporous Alumina

and Alumina-Titania Supported KCuFe Catalyst for Fischer-Tropsch Synthesis: Effects of

CO2 and CH4 Present in Syngas.” Fuel Processing Technology 168(August):140–51.

Baia, Luana V., Wallace C. Souza, Ricardo J. F. De Souza, Cláudia O. Veloso, Sandra S. X.

Chiaro, and Marco Antonio G. Figueiredo. 2017. “Removal of Sulfur and Nitrogen

Compounds from Diesel Oil by Adsorption Using Clays as Adsorbents.” Energy and Fuels

31(11):11731–42.

Bedda, K., Hamada, B., Semikin, K.V., Kuzichkin, N. V. 2019. “Petroleum and Coal

65

Desulfurization of Light Cycle Oil by Extraction with Polar Organic Solvents.” 61:1352–

60.

Biesinger, Mark C., Leo W. M. Lau, Andrea R. Gerson, and Roger St C. Smart. 2010.

“Resolving Surface Chemical States in XPS Analysis of First Row Transition Metals,

Oxides and Hydroxides: Sc, Ti, V, Cu and Zn.” Applied Surface Science 257(3):887–98.

Biswas, Piyali, Prabhu Narayanasarma, Chandra Mouli Kotikalapudi, Ajay K. Dalai, and John

Adjaye. 2011. “Characterization and Activity of ZrO2 Doped SBA-15 Supported NiMo

Catalysts for HDS and HDN of Bitumen Derived Heavy Gas Oil.” Industrial and

Engineering Chemistry Research 50(13):7882–95.

Boahene, Philip E., Kapil K. Soni, Ajay K. Dalai, and John Adjaye. 2013. “Hydroprocessing of

Heavy Gas Oils Using FeW/SBA-15 Catalysts: Experimentals, Optimization of Metals

Loading, and Kinetics Study.” Catalysis Today 207:101–11.

Bokade, Vijay V. and Ganapati D. Yadav. 2009. “Transesterification of Edible and Nonedible

Vegetable Oils with Alcohols over Heteropolyacids Supported on Acid-Treated Clay.”

Industrial and Engineering Chemistry Research 48(21):9408–15.

Botana de la Cruz, Anakaren, Philip Boahene, Sundaramurthy Vedachalam, Ajay K. Dalai, and

John Adjaye. 2020. “Adsorptive Desulfurization through Charge-Transfer Complex Using

Mesoporous Adsorbents.” Fuel 269(February):117379.

Bu, Jie, Gabriel Loh, Chuandayani Gunawan Gwie, Silvia Dewiyanti, Michael Tasrif, and

Armando Borgna. 2011. “Desulfurization of Diesel Fuels by Selective Adsorption on

Activated Carbons: Competitive Adsorption of Polycyclic Aromatic Sulfur Heterocycles

and Polycyclic Aromatic Hydrocarbons.” Chemical Engineering Journal 166(1):207–17.

Canada, Goverment of. n.d. “Sulphur-Gasoline @ Www.Canada.Ca.”

Chang, Hua and Zong Xin Wu. 2009. “Experimental Study on Adsorption of Carbon Dioxide By

5A Molecular Sieve for Helium Purification of High-Temperature Gas-Cooled Reactor.”

Industrial and Engineering Chemistry Research 48(9):4466–73.

Chiou, Ming Shen and Hsing Ya Li. 2002. “Equilibrium and Kinetic Modeling of Adsorption of

Reactive Dye on Cross-Linked Chitosan Beads.” Journal of Hazardous Materials

93(2):233–48.

Chitanda, Jackson M., Prachee Misra, Ali Abedi, Ajay K. Dalai, and John D. Adjaye. 2015.

“Synthesis and Characterization of Functionalized Poly(Glycidyl Methacrylate)-Based

Particles for the Selective Removal of Nitrogen Compounds from Light Gas Oil: Effect of

Linker Length.” Energy and Fuels 29(3):1881–91.

Duan, Feifei, Chaoqiu Chen, Guizhen Wang, Yongzhen Yang, Xuguang Liu, and Yong Qin.

2014. “Efficient Adsorptive Removal of Dibenzothiophene by Graphene Oxide-Based

Surface Molecularly Imprinted Polymer.” RSC Advances 4(3):1469–75.

Duprez, Daniel, Adam F. Lee, Karen Wilson, Sandeep Badoga, Rajesh V. Sharma, Ajay K.

Dalai, John Adjaye, S. K. Maity, J. Ancheyta, L. Soberanis, F. Alonso, M. E. Llanos,

Sandeep Badoga, Rajesh V. Sharma, Ajay K. Dalai, John Adjaye, Vahid Vosoughi,

Sandeep Badoga, Ajay K. Dalai, Nicolas Abatzoglou, John Adjaye, Yongfeng Hu, Jie Bu,

Gabriel Loh, Chuandayani Gunawan Gwie, Silvia Dewiyanti, Michael Tasrif, Armando

Borgna, Hong Yang, Jinwen Chen, Craig Fairbridge, Yevgenia Briker, Yu Jie Zhu, and

66

Zbigniew Ring. 2011. “Inhibition of Nitrogen Compounds on the Hydrodesulfurization of

Substituted Dibenzothiophenes in Light Cycle Oil.” Fuel Processing Technology

489(1):86–97.

Duprez, Daniel and Karen Wilson. 2009. “An Efficient Route to Highly Organized , Tunable

Macroporous - Mesoporous.” 12896–97.

Espantaleón, A. G., J. A. Nieto, M. Fernández, and A. Marsal. 2003. “Use of Activated Clays in

the Removal of Dyes and Surfactants from Tannery Waste Waters.” Applied Clay Science

24(1–2):105–10.

Favvas, Evangelos P., Nikolaos S. Heliopoulos, Sergios K. Papageorgiou, Athanasios Ch

Mitropoulos, George C. Kapantaidakis, and Nick K. Kanellopoulos. 2015. “Helium and

Hydrogen Selective Carbon Hollow Fiber Membranes: The Effect of Pyrolysis Isothermal

Time.” Separation and Purification Technology 142:176–81.

Fei, Liu, Jingwei Rui, Rijie Wang, Yanfei Lu, and Xiaoxia Yang. 2017. “Equilibrium and

Kinetic Studies on the Adsorption of Thiophene and Benzothiophene onto NiCeY Zeolites.”

RSC Advances 7(37):23011–20.

Fleming, Hubert L., P. Kenneth, and K. Emmanuel. 1988. “United States Patent ( 19 ) Patent

Number : 4762537.” (19).

Foo, K. Y. and B. H. Hameed. 2010. “Insights into the Modeling of Adsorption Isotherm

Systems.” Chemical Engineering Journal 156(1):2–10.

Fu, Jinmei, Geoffrey C. Klein, Donald F. Smith, Sunghwan Kim, Ryan P. Rodgers, Christopher

L. Hendrickson, and Alan G. Marshall. 2006. “Comprehensive Compositional Analysis of

Hydrotreated and Untreated Nitrogen-Concentrated Fractions Syncrude Oil by Electron

Ionization, Field Desorption Ionization, and Electrospray Ionization Ultrahigh-Resolution

FT-ICR Mass Spectrometry.” Energy and Fuels 20(3):1235–41.

Gonz, V. and E. Sastre. 2001. “Thermally Stable Mesoporous Alumina Synthesized with Non-

Ionic Surfactants in the Presence of Amines.” 45:203–10.

Guevara, A., A. Alvarez, and M. Vrinat. 2008. “Effect of TiO2-Al2O3sol-Gel Supports on the

Superficial Ni and Mo Species in Oxidized and Sulfided NiMo/TiO2-Al2O3catalysts:

Influence on Dibenzothiophene Hydrodesulfurization.” Catalysis Letters 126(3–4):268–74.

Ha, Jeong W., Tenzing Japhe, Tesfamichael Demeke, Bertin Moreno, and Abel E. Navarro.

2018. “Clean Technologies on the Removal and Desorption of Sulfur Compounds from

Model Fuels with Modified Clays.” 58–69.

Han, Xue, Hongfei Lin, and Ying Zheng. 2014. “Regeneration Methods to Restore Carbon

Adsorptive Capacity of Dibenzothiophene and Neutral Nitrogen Heteroaromatic

Compounds.” Chemical Engineering Journal 243:315–25.

Heilig, Morton L. 1994. “United States Patent Office.” Computer Graphics 28(2):131–34.

Hendricks, David. 2011. Water Treatment Unit Processes. Boca Raton: Taylor and Francis

Group, LLC.

Hentze, H. P. and M. Antonietti. 2002. “Porous Polymers and Resins for Biotechnological and

Biomedical Applications.” Reviews in Molecular Biotechnology 90(1):27–53.

Herbert, Javier, Víctor Santes, Maria Teresa Cortez, René Zárate, Leonardo Díaz, Ali Abedi,

67

Jackson M. Chitanda, Ajay K. Dalai, John Adjaye, M. A. Ahmed, Silvester Tursiloadi,

Hiroaki Imai, Hiroshi Hirashima, Jae Hyung Kim, Xiaoliang Ma, Anning Zhou, and

Chunshan Song. 2005. “Catalytic Hydrotreating of Heavy Gasoil FCC Feed over a NiMo/γ-

Al2O3-TiO2 Catalyst: Effect of Hydrogen Sulfide on the Activity.” Catalysis Today 111(1–

2):74–83.

Hernández-Maldonado, Arturo J. and Ralph T. Yang. 2003. “Desulfurization of Commercial

Liquid Fuels by Selective Adsorption via π-Complexation with Cu(I)-Y Zeolite.” Industrial

and Engineering Chemistry Research 42(13):3103–10.

Jayaraman, Ambalavanan, Frances H. Yang, and Ralph T. Yang. 2006. “Effects of Nitrogen

Compounds and Polyaromatic Hydrocarbons on Desulfurization of Liquid Fuels by

Adsorption via π-Complexation with Cu(I)Y Zeolite.” Energy and Fuels 20(3):909–14.

Kale, S., S. B. Umbarkar, M. K. Dongare, R. Eckelt, U. Armbruster, and A. Martin. 2015.

“Selective Formation of Triacetin by Glycerol Acetylation Using Acidic Ion-Exchange

Resins as Catalyst and Toluene as an Entrainer.” Applied Catalysis A: General 490:10–16.

Khodabandeh, Shervin and Mark E. Davis. 2014. “Synthesis of Pure Alumina Mesoporous

Materials.” Chemistry of Materials 4756(5):1451–64.

Kim, Jae Hyung, Xiaoliang Ma, Anning Zhou, and Chunshan Song. 2006. “Ultra-Deep

Desulfurization and Denitrogenation of Diesel Fuel by Selective Adsorption over Three

Different Adsorbents: A Study on Adsorptive Selectivity and Mechanism.” Catalysis Today

111(1–2):74–83.

Laidler, Keith J. 1984. “The Development of the Arrhenius Equation.” Journal of Chemical

Education 61(6):494–98.

Lazzaroni, R. and M. Hecq. 2003. “Growth of Ultrathin Ti Films Deposited on SnO 2 by

Magnetron Sputtering.” Thin Solid Films 437(03):57–62.

Lemaire, Marc, Michéle Monnet, Michel Vrinat, Valérie Lamure, Emmanuelle Sanson, and

Alexandra Milenkovic. 2002. “Method for Separating Benzothiophene Compounds from

Hydrocarbon Mixture Containing Them, and Hydrocarbon Mixture Obtained by Said

Method.” US. PATENT DOCUMENTS 1(12):1–5.

Li, Chong Jiu, Ran Zhao, Meng Qi Peng, Hao Liu, Gang Yu, and Dong Sheng Xia. 2015. “Study

on Desulfurization and Denitrification by Modified Activated Carbon Fibers with Visible-

Light Photocatalysis.” Ranliao Huaxue Xuebao/Journal of Fuel Chemistry and Technology

43(12):1516–22.

Li, Gang, Penny Xiao, Paul Webley, Jun Zhang, Ranjeet Singh, and Marc Marshall. 2008.

“Capture of CO2 from High Humidity Flue Gas by Vacuum Swing Adsorption with Zeolite

13X.” Adsorption 14(2–3):415–22.

Li, Pam H. Y., Roderick L. Bruce, and Malcolm D. Hobday. 1999. “A Pseudo First Order Rate

Model for the Adsorption of an Organic Adsorbate in Aqueous Solution.” Journal of

Chemical Technology and Biotechnology 74(1):55–59.

Li, Weiwei, Jingwen Chen, Guojing Cong, Lei Tang, Qun Cui, and Haiyan Wang. 2016.

“Solvent Desulfurization Regeneration Process and Analysis of Activated Carbon for Low-

Sulfur Real Diesel.” RSC Advances 6(24):20258–68.

Li, Xiaojuan, Xingwang Zhang, and Lecheng Lei. 2009. “Preparation of CuNaY Zeolites with

68

Microwave Irradiation and Their Application for Removing Thiophene from Model Fuel.”

Separation and Purification Technology 64(3):326–31.

Li, Y. Y., S. P. Perera, and B. D. Crittenden. 1998. “Zeolite Monoliths for Air Separation Part 2:

Oxygen Enrichment, Pressure Drop and Pressurization.” Chemical Engineering Research

and Design 76(8 A8):931–41.

Lin, J. X., S. L. Zhan, M. H. Fang, and X. Q. Qian. 2007. “The Adsorption of Dyes from

Aqueous Solution Using Diatomite.” Journal of Porous Materials 14(4):449–55.

Lin, Junxiong and Lan Wang. 2009. “Comparison between Linear and Non-Linear Forms of

Pseudo-First-Order and Pseudo-Second-Order Adsorption Kinetic Models for the Removal

of Methylene Blue by Activated Carbon.” Frontiers of Environmental Science and

Engineering in China 3(3):320–24.

Liu, Bing, Xiaochen Xu, Yuan Xue, Lifen Liu, and Fenglin Yang. 2020. “Simultaneous

Desulfurization and Denitrification from Flue Gas by Catalytic Ozonation Combined with

NH3/(NH4)2S2O8 Absorption: Mechanisms and Recovery of Compound Fertilizer.”

Science of the Total Environment 706:136027.

Macaud, Mathieu, Marc Sévignon, Alain Favre-Réguillon, Marc Lemaire, Emmanuelle Schulz,

and Michel Vrinat. 2004. “Novel Methodology toward Deep Desulfurization of Diesel Feed

Based on the Selective Elimination of Nitrogen Compounds.” Industrial & Engineering

Chemistry Research 43(24):7843–49.

Maity, S. K., J. Ancheyta, Mohan S. Rana, and P. Rayo. 2006. “Alumina - Titania Mixed Oxide

Used as Support for Hydrotreating Catalysts of Maya Heavy Crude - Effect of Support

Preparation Methods.” Energy and Fuels 20(2):427–31.

Maity, S. K., J. Ancheyta, L. Soberanis, F. Alonso, and M. E. Llanos. 2003. “Alumina-Titania

Binary Mixed Oxide Used as Support of Catalysts for Hydrotreating of Maya Heavy

Crude.” Applied Catalysis A: General 244(1):141–53.

Malik, P. K. 2004. “Dye Removal from Wastewater Using Activated Carbon Developed from

Sawdust: Adsorption Equilibrium and Kinetics.” Journal of Hazardous Materials 113(1–

3):81–88.

Mccabe, Peter J. 2020. “Fossil Energy.” Fossil Energy.

Mckav, G. and Northern Ireland. 1985. “The Adsorption of Various Pollutants from Aqueous

Solutions onto Activated Carbon.” Water Research 19(4):491–95.

McKay, G., M. J. Bino, and A. R. Altamemi. 1985. “The Adsorption of Various Pollutants from

Aqueous Solutions on to Activated Carbon.” Water Research 19(4):491–95.

Meille, Valérie, Emmanuelle Schulz, Michel Vrinat, and Marc Lemaire. 1998. “A New Route

towards Deep Desulfurization: Selective Charge Transfer Complex Formation.” Chemical

Communications (3):305–6.

Michard, P., E. Guibal, T. Vincent, and P. Le Cloirec. 1996. “Sorption and Desorption of Uranyl

Ions by Silica Gel: PH, Particle Size and Porosity Effects.” Microporous Materials

5(5):309–24.

Milenkovic, Alexandra, David Loffreda, Emmanuelle Schulz, Henry Chermette, De Lyon, Paris

Sud, Paris Xi, and Orsay Cedex. 2004. “And Experimental Study.” 1169–80.

69

Milenkovic, Alexandra, Emmanuelle Schulz, Valérie Meille, David Loffreda, Michel Forissier,

Michel Vrinat, Philippe Sautet, and Marc Lemaire. 1999. “Selective Elimination of

Alkyldibenzothiophenes from Gas Oil by Formation of Insoluble Charge-Transfer

Complexes.” Energy and Fuels 13(4):881–87.

Misra, Prachee. 2017. “Desulfurization and Denitrogenation of Bitumen-Derived Gas Oils Using

Functionalized Polymers.” (May):1–164.

Misra, Prachee, Sandeep Badoga, Ajay K. Dalai, and John Adjaye. 2018. “Enhancement of

Sulfur and Nitrogen Removal from Heavy Gas Oil by Using Polymeric Adsorbent Followed

by Hydrotreatment.” Fuel 226(April):127–36.

Misra, Prachee, Jackson M. Chitanda, Ajay K. Dalai, and John Adjaye. 2015a. “Immobilization

of Fluorenone Derived π-Acceptors on Poly (GMA-Co-EGDMA) for the Removal of

Refractory Nitrogen Species from Bitumen Derived Gas Oil.” Fuel 145:100–108.

Misra, Prachee, Jackson M. Chitanda, Ajay K. Dalai, and John Adjaye. 2015b. “Immobilization

of Fluorenone Derived π-Acceptors on Poly (GMA-Co-EGDMA) for the Removal of

Refractory Nitrogen Species from Bitumen Derived Gas Oil.” Fuel 145:100–108.

Misra, Prachee, Jackson M. Chitanda, Ajay K. Dalai, and John Adjaye. 2016. “Selective

Removal of Nitrogen Compounds from Gas Oil Using Functionalized Polymeric

Adsorbents: Efficient Approach towards Improving Denitrogenation of Petroleum

Feedstock.” Chemical Engineering Journal 295:109–18.

Mochida, Isao, Yozo Korai, Masuaki Shirahama, Shizuo Kawano, Tomohiro Hada, Yorimasa

Seo, Masaaki Yoshikawa, and Akinori Yasutake. 2000. “Removal of SOx and NOx over

Activated Carbon Fibers.” Carbon 38(2):227–39.

Munakata, Kenzo, Teruki Fukumatsu, Satoshi Yamatsuki, and Kenji Tanaka. 1999. “Adsorption

Equilibria of Krypton, Xenon, Nitrogen and Their Mixtures on Molecular Sieve 5A and

Activated Charcoal.” Journal of Nuclear Science and Technology 36(9):818–29.

Nathan, Andrew J. and Andrew Scobell. 2012. “How China Sees America.” Foreign Affairs

91(5):2.1-2.54.

Niesz, Krisztian, Peidong Yang, and Gabor A Somorjai. 2005. “Sol-Gel Synthesis of Ordered

Mesoporous Alumina.” RSC Advances 1986–87.

Niesz, Krisztian, Peidong Yang, and Gabor A. Somorjai. 2005. “Sol-Gel Synthesis of Ordered

Mesoporous Alumina.” Chemical Communications (15):1986–87.

Pan, Bo and Baoshan Xing. 2010. “Adsorption Kinetics of 17α-Ethinyl Estradiol and Bisphenol

A on Carbon Nanomaterials. I. Several Concerns Regarding Pseudo-First Order and

Pseudo-Second Order Models.” Journal of Soils and Sediments 10(5):838–44.

Pouilleau, J., D. Devilliers, F. Garrido, S. Durand-vidal, and E. Mah. 1997. “Structure and

Composition of Passive Titanium Oxide Films.” Materials Science and Engineering

B47:235–43.

Puoci, Francesco Iemma, Francesca Gianfranco, Unile Cirillo, Giuseppe. 2008. “Polymer in

Agriculture: A Review.” American Journal of Agricultural and Biological Sciences

299(314).

El Qada, Emad N., Stephen J. Allen, and Gavin M. Walker. 2006. “Adsorption of Methylene

70

Blue onto Activated Carbon Produced from Steam Activated Bituminous Coal: A Study of

Equilibrium Adsorption Isotherm.” Chemical Engineering Journal 124(1–3):103–10.

Qi, Liqiang, Zhikai Zhao, Ruitao Wang, Weiheng Gao, Jingxin Li, and Yajuan Zhang. 2020.

“Simultaneous Desulfurization and Denitri Fi Cation Using La − Ce − V − Cu − ZSM ‑ 5

Catalysts in an Electrostatic Precipitator.”

Rendon-Rivera, A., M. A. Cortes-Jacome, E. Lopez-Salinas, M. L. Mosqueira, and J. A. Toledo-

Antonio. 2016. “Adsorption of Nitrogen and Sulphur Organic-Compounds on Titania

Nanotubes.” International Journal of Engineering Research & Science 2(8):156–66.

Richardson, Louis L. 1978. “Use of Bleaching, Clays, in Processing Edible Oils.” Journal of the

American Oil Chemists’ Society 55(11):777–80.

Rizwan, Danish, Ajay K. Dalai, and John Adjaye. 2013. “Synthesis of Novel Polymer

Poly(Glycidyl Methacrylate) Incorporated with Tetranitrofluorenone for Selective Removal

of Neutral Nitrogen Species from Bitumen-Derived Heavy Gas Oil.” Fuel Processing

Technology 106:483–89.

Ronald F. Ziolo, Webster, N. Y. .., Canada; Elizabeth C. Kroll, Hamilton, Javier Tejada Palacios;

Xixiang, and Spain Zhang, both of Barcelona. 1995. “Magnetc Refrgerant Compostions and

Processes for Makng and Usng.” 11.

Sampanthar, Jeyagowry T., Huang Xiao, Jian Dou, Teo Yin Nah, Xu Rong, and Wong Pui

Kwan. 2006. “A Novel Oxidative Desulfurization Process to Remove Refractory Sulfur

Compounds from Diesel Fuel.” Applied Catalysis B: Environmental 63(1–2):85–93.

Sano, Yosuke, Ki Hyouk Choi, Yozo Korai, and Isao Mochida. 2004. “Selection and Further

Activation of Activated Carbons for Removal of Nitrogen Species in Gas Oil as a

Pretreatment for Its Deep Hydrodesulfurization.” Energy and Fuels 18(3):644–51.

Santos, Aaron L., Rodrigo A. Reis, Vinicius Rossa, Marcelo M. Reis, André L. H. Costa,

Cláudia O. Veloso, Cristiane A. Henriques, Fátima M. Z. Zotin, Márcio L. L. Paredes, Erika

B. Silveira, and Sandra S. X. Chiaro. 2012. “Silica-Alumina Impregnated with Cerium,

Nickel, and Molybdenum Oxides for Adsorption of Sulfur and Nitrogen Compounds from

Diesel.” Materials Letters 83:158–60.

Sarda, K. K., A. Bhandari, K. K. Pant, and Sapna Jain. 2012. “Deep Desulfurization of Diesel

Fuel by Selective Adsorption over Ni/Al 2O 3 and Ni/ZSM-5 Extrudates.” Fuel 93:86–91.

Satterfield, Charles N. 1996. Heterogeneous Catalysis in Industrial Practice 2nd E. 2nd Editio.

Krieger Publishing Company.

Schirmer, W. 1999. “Physical Chemistry of Surfaces.” Zeitschrift Für Physikalische Chemie

210(Part_1):134–35.

Sévignon, Marc, Mathieu Macaud, Alain Favre-Réguillon, Jürgen Schulz, Muriel Rocault, René

Faure, Michel Vrinat, and Marc Lemaire. 2005. “Ultra-Deep Desulfurization of

Transportation Fuels via Charge-Transfer Complexes under Ambient Conditions.” Green

Chemistry 7(6):413–20.

Shan, Jia Hui, Xiao Qin Liu, Lin Bing Sun, and Rong Cui. 2008. “Cu-Ce Bimetal Ion-Exchanged

Y Zeolites for Selective Adsorption of Thiophenic Sulfur.” Energy and Fuels 22(6):3955–

59.

71

Silva, Alysson M. A., Eduardo H. M. Nunes, Douglas F. Souza, Dana L. Martens, João C. Diniz

da Costa, Manuel Houmard, and Wander L. Vasconcelos. 2015. “Effect of Titania Addition

on the Properties of Freeze-Cast Alumina Samples.” Ceramics International 41:10467–75.

Silveira, E., C. Veloso, A. Costa, C. Henriques, F. Zotin, M. Paredes, R. Reis, and S. Chiaro.

2015. “Influence of Metal Oxides Impregnated on Silica-Alumina in the Removal of

Sulphur and Nitrogen Compounds from a Hydrotreated Diesel Fuel Stream.” Adsorption

Science and Technology 33(2):105–16.

Sismanoglu, T. and S. Pura. 2001. “Adsorption of Aqueous Nitrophenols on Clinoptilolite.”

Colloids and Surfaces A: Physicochemical and Engineering Aspects 180(1–2):1–6.

Song, Chunshan. 2003. “An Overview of New Approaches to Deep Desulfurization for Ultra-

Clean Gasoline, Diesel Fuel and Jet Fuel.” Catalysis Today 86(1–4):211–63.

Speight, James G. 2020. “Fossil Energy.” Fossil Energy 17–45.

Sumathi, S., S. Bhatia, K. T. Lee, and A. R. Mohamed. 2009. “Performance of an Activated

Carbon Made from Waste Palm Shell in Simultaneous Adsorption of so x and NO x of Flue

Gas at Low Temperature.” Science in China, Series E: Technological Sciences 52(1):198–

203.

Sumathi, S., S. Bhatia, K. T. Lee, and A. R. Mohamed. 2010. “Selection of Best Impregnated

Palm Shell Activated Carbon (PSAC) for Simultaneous Removal of SO2 and NOx.”

Journal of Hazardous Materials 176(1–3):1093–96.

Takahashi, Akira, Frances H. Yang, and Ralph T. Yang. 2002. “New Sorbents for

Desulfurization by π -Complexation : Thiophene / Benzene Adsorption.” 2487–96.

Tang, Hu, Weijie Zhou, and Lina Zhang. 2012. “Adsorption Isotherms and Kinetics Studies of

Malachite Green on Chitin Hydrogels.” Journal of Hazardous Materials 209–210:218–25.

Toor, Saqib S., Harvind Reddy, Shuguang Deng, Jessica Hoffmann, Dorte Spangsmark, Linda B.

Madsen, Jens Bo Holm-Nielsen, and Lasse A. Rosendahl. 2013. “Hydrothermal

Liquefaction of Spirulina and Nannochloropsis Salina under Subcritical and Supercritical

Water Conditions.” Bioresource Technology 131:413–19.

Tursiloadi, Silvester, Hiroaki Imai, and Hiroshi Hirashima. 2004. “Preparation and

Characterization of Mesoporous Titania-Alumina Ceramic by Modified Sol-Gel Method.”

Journal of Non-Crystalline Solids 350:271–76.

Ullah, K. R., R. Saidur, H. W. Ping, R. K. Akikur, and N. H. Shuvo. 2013. “A Review of Solar

Thermal Refrigeration and Cooling Methods.” Renewable and Sustainable Energy Reviews

24:499–513.

Vaez, Mohammad, Abdolsamad Zarringhalam Moghaddam, and Somayeh Alijani. 2012.

“Optimization and Modeling of Photocatalytic Degradation of Azo Dye Using a Response

Surface Methodology (RSM) Based on the Central Composite Design with Immobilized

Titania Nanoparticles.” Industrial and Engineering Chemistry Research 51(11):4199–4207.

Velu, S., Chunshan Song, Clean Fuels, Catalysis Program, and Geo-environmental Engineering.

2002. “Zeolite-Based Adsorbents for Desulfurization of Jet Fuel by Selective Adsorption.”

Fuel Chemistry 47(2):447–48.

Vosoughi, Vahid, Sandeep Badoga, Ajay K. Dalai, and Nicolas Abatzoglou. 2017. “Modification

72

of Mesoporous Alumina as a Support for Cobalt-Based Catalyst in Fischer-Tropsch

Synthesis.” Fuel Processing Technology 162:55–65.

Vural, Ufuk. 2020. “Waste Mineral Oils Re-Refining With Physicochemical Methods.” Turkish

Journal of Engineering 4(2):62–69.

Wachs, Israel E. 2005. “Recent Conceptual Advances in the Catalysis Science of Mixed Metal

Oxide Catalytic Materials.” Catalysis Today 100(1–2):79–94.

Weber, Rainer, Hans-dieter Block, and Norbert Lonhoff. 1991. “United States Patent [ 151 Date

of Patent :” (19):3–5.

Wu, Xiaohui, Bo Yue, Yi Su, Qi Wang, Qifei Huang, Qunhui Wang, and Hongying Cai. 2017.

“Pollution Characteristics of Polycyclic Aromatic Hydrocarbons in Common Used Mineral

Oils and Their Transformation during Oil Regeneration.” Journal of Environmental

Sciences (China) 56(200 L):247–53.

Yang, Hong, Jinwen Chen, Craig Fairbridge, Yevgenia Briker, Yu Jie Zhu, and Zbigniew Ring.

2004. “Inhibition of Nitrogen Compounds on the Hydrodesulfurization of Substituted

Dibenzothiophenes in Light Cycle Oil.” Fuel Processing Technology 85(12):1415–29.

Yang, Ralph T., Akira Takahashi, and Frances H. Yang. 2001. “New Sorbents for

Desulfurization of Liquid Fuels by π -Complexation.” 6236–39.

Yang, Yuqing, Shongming Huang, Robert Vassov, Brad Pinno, and Sophan Chhin. 2020.

“Climate-Sensitive Height – Age Models for Top Height Trees in Natural and Reclaimed

Oil Sands Stands in Alberta , Canada.” 307(November 2019):297–307.

Yitzhaki, D. 1995. “Deep Desulfurization of Heavy Atmospheric Gas Oil With Co-Mo-Al

Catalysts - Effect of Sulfur Adsorption.” Applied Catalysis A-General 122(2):99–110.

Yoosuk, Boonyawan, Arnan Silajan, and Pattarapan Prasassarakich. 2020. “Deep Adsorptive

Desulfurization over Ion-Exchanged Zeolites: Individual and Simultaneous Effect of

Aromatic and Nitrogen Compounds.” Journal of Cleaner Production 248:119291.

Yuan, Quan, An-Xiang Yin, Chen Luo, Ling-Dong Sun, Ya-Wen Zhang, Wen-Tao Duan, Hai-

Chao Liu, and Chun-Hua Yan. 2008. “Facile Synthesis for Ordered Mesoporous γ-

Aluminas with High Thermal Stability.” Journal of the American Chemical Society

130(11):3465–72.

Zhang, Yong, Feng Yu, Wenping Cheng, Jiancheng Wang, and Juanjuan Ma. 2017. “Adsorption

Equilibrium and Kinetics of the Removal of Ammoniacal Nitrogen by Zeolite X/Activated

Carbon Composite Synthesized from Elutrilithe.” Journal of Chemistry 2017.

Zhou, Keqing, Qiangjun Zhang, Biao Wang, Jiajia Liu, Panyue Wen, Zhou Gui, and Yuan Hu.

2014. “The Integrated Utilization of Typical Clays in Removal of Organic Dyes and

Polymer Nanocomposites.” Journal of Cleaner Production 81:281–89.

Zolotareva, Darya, Alexey Zazybin, Khadichakhan Rafikova, Valery M. Dembitsky, Anuar

Dauletbakov, and Valentina Yu. 2019. “Ionic Liquids Assisted Desulfurization and

Denitrogenation of Fuels.” Vietnam Journal of Chemistry 57(2):133–63.

73

Appendix A: Permission to reuse submitted paper Tables and Figures

Figure A.1.Permission to use published article, tables and Figures for Chapter 3

74

Appendix B: Additional table of results from process parameter optimization

studies and activation energy (Ea) plots for adsorbents A, B and C

Figure B.1. Optimization result table for adsorbent A

75

Figure B.2. Optimization result table for adsorbent B

76

Figure B.3. Optimization result table for adsorbent C

Figure B.4. Activation energy (Ea) plot ln KD vs. 1/T (K-1) for adsorbent A

y = 963.65x - 7.1746

R² = 0.97

-4.4

-4.3

-4.2

-4.1

-4

-3.9

-3.8

-3.7

-3.6

-3.5

0.00275 0.003 0.00325 0.0035

ln K

D

1/T (K)

Adsorbent A

77

Figure B.5. Activation energy (Ea) plot ln KD vs. 1/T (K-1) for adsorbent B

Figure B.6. Activation energy (Ea) plot ln KD vs. 1/T (K-1) for adsorbent C

y = -1531.6x + 0.0619

R² = 0.37

-5.40

-5.20

-5.00

-4.80

-4.60

-4.40

-4.20

-4.00

0.0028 0.0030 0.0033 0.0035

ln K

D

1/T (K-1)

Adsorbent B

y = 958.97x - 6.2779

R² = 0.99

-3.5

-3.4

-3.3

-3.2

-3.1

-3

-2.9

0.0029 0.0030 0.0031 0.0032 0.0033 0.0034 0.0035

ln K

D

1/T (K-1)

Adsorbent C

78

Figure B.7. Free activation energy, enthalpy and entropy plots for adsorbents A, B and C

y = 963.65x - 7.1746

R² = 0.97

y = -1531.6x + 0.0619

R² = 0.37

y = 958.97x - 6.2779

R² = 0.99

-5.5

-5

-4.5

-4

-3.5

-3

-2.5

-2

2.80E-03 3.20E-03 3.60E-03

ln K

D

1/T (K-1)

Adsorbent A

Adsorbent B